EP4706017A2 - Systems and methods for inline quality control of slide digitization - Google Patents

Abstract

Described herein is a system for digitizing a slide. A system may include an optical system including at least an optical sensor; and a computing device configured to perform a scan of the slide by capturing at least a first image and capturing at least a second image, wherein performing the scan includes identifying a first scanning parameter; using the optical system, capturing the at least a first image as a function of the first scanning parameter; determining a first quality metric as a function of the at least a first image; determining a second scanning parameter as a function of the first quality metric; and using the optical system, capturing the at least a second image as a function of the second scanning parameter.

Description

SYSTEMS AND METHODS FOR INLINE QUALITY CONTROL OF SLIDE DIGITIZATION CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit of priority of U.S. Nonprovisional Application Serial No.18/602,947, filed on March 12, 2024, and entitled “SYSTEMS AND METHODS FOR INLINE QUALITY CONTROL OF SLIDE DIGITIZATION”, which claims the benefit of priority of U.S. Provisional Application Serial No.63/466,950, filed on May 16, 2023, and entitled “SYSTEMS AND METHODS FOR INLINE QUALITY CONTROL OF SLIDE DIGITIZATION”, each of which is incorporated by reference herein in its entirety. This application also claims the benefit of priority of U.S. Nonprovisional Application Serial No. 18/603,051, filed on March 12, 2024, and entitled “SYSTEMS AND METHODS FOR DIGITIZATION OF TISSUE SLIDES BASED ON ASSOCIATIONS AMONG SERIAL SECTIONS”, which claims the benefit of priority of U.S. Provisional Application Serial No. 63/465,032, filed on May 9, 2023, and entitled “SYSTEMS AND METHODS FOR DIGITIZATION OF TISSUE SLIDES BASED ON ASSOCIATIONS AMONG SERIAL SECTIONS”, each of which is incorporated by reference herein in its entirety. FIELD OF THE INVENTION The present invention generally relates to the field of slide digitization. In particular, the present invention is directed to techniques for inline quality control of slide digitization. BACKGROUND Examination of slides containing biomedical specimens, such as tissue samples, under a microscope provides data that can be exploited for a variety of biomedical applications. For example, physicians or other qualified individuals may be able to diagnose pathological conditions or detect microbial organisms. In many instances, the physician may observe the slides directly under the microscope. Increasingly, however, it is desirable to digitize microscope slides for downstream analysis. SUMMARY OF THE DISCLOSURE In an aspect, a system for digitizing a slide may include an optical system comprising at least an optical sensor; and a computing device configured to perform a scan of the slide by capturing at least a first image and capturing at least a second image, wherein performing the scan comprises identifying a first scanning parameter; using the optical system, capturing the at Attorney Docket No.1519-029PCT1 least a first image as a function of the first scanning parameter; determining a first quality metric as a function of the at least a first image; determining a second scanning parameter as a function of the first quality metric; and using the optical system, capturing the at least a second image as a function of the second scanning parameter. In some embodiments, the first quality metric comprises a localization quality metric; and determining the second scanning parameter comprises identifying at least a region of interest of the slide as a function of the localization quality metric. In some embodiments, the first quality metric comprises a biopsy plane estimation quality metric; and determining the second scanning parameter comprises identifying a plane within a region of interest which contains a biological specimen as a function of the biopsy plane estimation quality metric. In some embodiments, the first quality metric comprises a focus sampling quality metric; and determining the second scanning parameter comprises identifying a focal setting at which a selected point is in focus as a function of the focus sampling quality metric. In some embodiments, the first quality metric comprises a z-stack acquisition quality metric; and determining the second scanning parameter comprises, using the optical system, capturing a z-stack at a selected point as a function of the z-stack acquisition quality metric. In some embodiments, the at least a first image comprises a macro image; and capturing the at least a second image comprises using the optical system, capturing a first plurality of images; and combining the first plurality of images to create the at least a second image. In some embodiments, performing the scan further comprises determining a stitching quality metric as a function of the at least a second image; determining a third scanning parameter as a function of the stitching quality metric; using the optical system, capturing a second plurality of images as a function of the third scanning parameter; and combining the second plurality of images to create at least a third image. In some embodiments, performing the scan further comprises capturing the at least a first image at a first location and capturing the at least a second image at a second location. In some embodiments, performing the scan further comprises determining a second quality metric as a function of the at least a first image; determining a combination quality metric as a function of the first quality metric and the second quality metric; determining a third scanning parameter as a function of the combination quality metric; and using the optical system, capturing at least a third image as a function of the third scanning parameter. In some embodiments, performing the scan further comprises algorithmically removing a banding error from the at least a second image. 2 Attorney Docket No.1519-029PCT1 In another aspect, a method of digitizing a slide may include using at least a processor, performing a scan of the slide by capturing at least a first image and capturing at least a second image, wherein performing the scan comprises identifying a first scanning parameter; using an optical system, capturing the at least a first image as a function of the first scanning parameter; determining a first quality metric as a function of the at least a first image; determining a second scanning parameter as a function of the first quality metric; and using the optical system, capturing the at least a second image as a function of the second scanning parameter. In some embodiments, the first quality metric comprises a localization quality metric; and determining the second scanning parameter comprises identifying at least a region of interest of the slide as a function of the localization quality metric. In some embodiments, the first quality metric comprises a biopsy plane estimation quality metric; and determining the second scanning parameter comprises identifying a plane within a region of interest which contains a biological specimen as a function of the biopsy plane estimation quality metric. In some embodiments, the first quality metric comprises a focus sampling quality metric; and determining the second scanning parameter comprises identifying a focal setting at which a selected point is in focus as a function of the focus sampling quality metric. In some embodiments, the first quality metric comprises a z-stack acquisition quality metric; and determining the second scanning parameter comprises, using the optical system, capturing a z-stack at a selected point as a function of the z- stack acquisition quality metric. In some embodiments, the at least a first image comprises a macro image; and capturing the at least a second image comprises using the optical system, capturing a first plurality of images; and combining the first plurality of images to create the at least a second image. In some embodiments, performing the scan further comprises determining a stitching quality metric as a function of the at least a second image; determining a third scanning parameter as a function of the stitching quality metric; using the optical system, capturing a second plurality of images as a function of the third scanning parameter; and combining the second plurality of images to create at least a third image. In some embodiments, performing the scan further comprises capturing the at least a first image at a first location and capturing the at least a second image at a second location. In some embodiments, performing the scan further comprises determining a second quality metric as a function of the at least a first image; determining a combination quality metric as a function of the first quality metric and the second quality metric; determining a third scanning parameter as a function of the combination 3 Attorney Docket No.1519-029PCT1 quality metric; and using the optical system, capturing at least a third image as a function of the third scanning parameter. In some embodiments, performing the scan further comprises algorithmically removing a banding error from the at least a second image. A method comprising evaluating an inline quality metric associated with digitizing a slide; and in response to determining that the inline quality metric is within a predetermined range, taking a remedial action. In another aspect, a system for digitization of tissue slides based on associations among serial sections may include at least a computing device, wherein the computing device is comprised of a memory, wherein the memory stores instructions; and a processor, communicatively connected to the memory, wherein the processor is configured to retrieve a candidate tissue map associated with a candidate tissue section; retrieve a reference tissue map associated with a reference tissue section; align the candidate tissue map to the reference tissue map; compare the aligned candidate tissue map to the reference tissue map; and generate a regenerated candidate tissue map as a function of the reference tissue map; and a scanner, configured to scan a slide and send a digitized image of the slide to the computing device. In some embodiments, the memory further includes instructions configuring the processor to identify a candidate serial section from at least one stain type, wherein the candidate serial section is associated with a case identification number and a block identification number; identify a reference serial section based on the cane identification number and the block identification number; generate the reference tissue map in response to scanning the reference serial section; and generate the candidate tissue map in response to scanning the candidate serial section. In some embodiments, the candidate serial section is on a first slide and the reference serial section is on a second slide. In some embodiments, the candidate serial section and the reference serial section are both on a first slide. In some embodiments, the system is further comprised of at least a storage device. In some embodiments, the system is further comprised of a scanned slides data repository. In some embodiments, the system is further comprised of a regenerated slides data repository. In some embodiments, the system is further comprised of both a scanned slides data repository and a regenerated slides data repository. In some embodiments, the system instantiates a machine learning module. In some embodiments, the system instantiates a neural network. 4 Attorney Docket No.1519-029PCT1 In another aspect, a method for digitization of tissue slides based on associations among serial sections may include receiving a candidate tissue map associated with a candidate tissue section; receiving a reference tissue map associated with a reference tissue section, the reference tissue section having a predetermined association with the candidate tissue section; aligning the candidate tissue map to the reference tissue map; comparing the aligned candidate tissue map to the reference tissue map; and generating a regenerated candidate tissue map as a function of the reference tissue map. In some embodiments, the method further comprises identifying a candidate serial section from at least one stain type, wherein the candidate serial section is associated with a case identification number and a block identification number; identifying a reference serial section based on the case identification number and the block identification number; generating the reference tissue map in response to scanning the reference serial section; and generating the candidate tissue map in response to scanning the candidate serial section. In some embodiments, the candidate serial section is on a first slide and the reference serial section is on a second slide. In some embodiments, the candidate serial section and the reference serial section are both on a first slide. In some embodiments, the method further includes storage and retrieval of slides data from at least a storage device. In some embodiments, the method further includes storage and retrieval of slides data from a scanned slides data repository. In some embodiments, the method further includes storage and retrieval of slides data from a regenerated slides data repository. In some embodiments, the method further includes storage and retrieval of slides data from both a scanned slides data repository and a regenerated slides data repository. In some embodiments, the method instantiates a machine learning module. In some embodiments, the method instantiates a neural network. In another aspect, a slide digitization method may include receiving a candidate tissue map associated with a candidate tissue section; receiving a reference tissue map associated with a reference tissue section, the reference tissue section having a predetermined association with the candidate tissue section; aligning, by the processor, the candidate tissue map to the reference tissue map; comparing, by the processor, the aligned candidate tissue map to the reference tissue map; and in response to comparing the aligned candidate tissue map to the reference tissue map, generating, by the processor, a regenerated candidate tissue map. In some embodiments, a method may further include identifying, based on at least one stain type, a candidate serial section, the candidate serial section associated with a case identification number and a block 5 Attorney Docket No.1519-029PCT1 identification number; identifying, based on the case identification number and the block identification number, a reference serial section; generating, by a processor, the reference tissue map in response to scanning the reference serial section; and generating, by the processor, the candidate tissue map in response to scanning the candidate serial section. In some embodiments, the candidate serial section is on a first slide and the reference serial section is on a second slide. In some embodiments, the candidate serial section and the reference serial section are both on a first slide. A slide digitization system comprising a computing device, the computing device having a processor and a memory device in communication with the processor; and a scanner configured to scan a slide and send a digitized image of the slide to the computing device, wherein the processor, when executing instructions stored in the memory device, is configured to retrieve a candidate tissue map associated with a candidate tissue section; retrieve a reference tissue map associated with a reference tissue section, the reference tissue section having a predetermined association with the candidate tissue section; align the candidate tissue map to the reference tissue map; compare the aligned candidate tissue map to the reference tissue map; and in response to comparing the aligned candidate tissue map to the reference tissue map, generate a regenerated candidate tissue map. In some embodiments, the processor is further configured to generate a reference tissue map in response to a scanned reference serial section, the scanned reference serial section having been scanned by the scanner from a reference serial section; and generate a candidate tissue map in response to a scanned candidate serial section, the scanned candidate serial section having been scanned by the scanner from a candidate serial section. In some embodiments, the candidate serial section is on a first slide and the reference serial section is on a second slide. In some embodiments, the candidate serial section and the reference serial section are both on a first slide. In another aspect, an apparatus for inline image enrichment may include circuitry configured to receive a plurality of subject data corresponding to a subject; generate a model candidate set using the plurality of subject data; instantiate at least a model of the model candidate set; digitally capture an image of the subject, wherein digitally capturing the image comprises; identifying at least a focal point using the at least a model; and capturing the image as a function of the at least a focal point; and store the captured image in a repository. In some embodiments, the circuit includes a configurable hardware circuit. In some embodiments, generating the candidate model set further comprises classifying the plurality of subject data to at 6 Attorney Docket No.1519-029PCT1 least a candidate model of a plurality of potential candidate models. In some embodiments, the plurality of subject data includes at least an element of subject metadata; and generating the candidate model set further comprises generating the candidate model set using the subject metadata. In some embodiments, instantiating the at least a model further comprises instantiating a locally cached model. In some embodiments, instantiating the at least a model further comprises receiving a remotely cached model; and instantiating the remotely cached model. In some embodiments, capturing the image further comprises reviewing an initial scan; comparing the initial scan to a confidence threshold; and performing a subsequent scan based on the comparison. In some embodiments, capturing the image further comprises capturing a plurality of individual stack views of the image; saving each individual stack view; and executing a continuum diffusion process to fuse individual stacks together. In some embodiments, the plurality of individual stack views correspond to a plurality of distinct focal points. In some embodiments, an apparatus may be further configured to enhance at least a viewability characteristic of the image using a machine-learning process. In another aspect, a method for inline image enrichment may include receiving, by the configured circuitry, a plurality of subject data corresponding to a subject; generating, by the configured circuitry, a model candidate set using the plurality of subject data; instantiating, by the configured circuitry, at least a model of the model candidate set; digitally capturing, by the configured circuitry, an image of the subject, wherein digitally capturing the image comprises; identifying, by the configured circuitry, at least a focal point using the at least a model; and capturing, by the configured circuitry, the image as a function of the at least a focal point; and storing, by the configured circuitry, the captured image in a repository. In some embodiments, the circuit includes a configurable hardware circuit. In some embodiments, generating the candidate model set further comprises classifying, by the configured circuitry, the plurality of subject data to at least a candidate model of a plurality of potential candidate models. In some embodiments, the plurality of subject data includes at least an element of subject metadata; and generating the candidate model set further comprises generating, by the configured circuitry, the candidate model set using the subject metadata. In some embodiments, instantiating the at least a model further comprises instantiating, by the configured circuitry, a locally cached model. In some embodiments, instantiating the at least a model further comprises receiving, by the configured circuitry, a remotely cached model; and instantiating the remotely cached model. In 7 Attorney Docket No.1519-029PCT1 some embodiments, capturing the image further comprises reviewing, by the configured circuitry, an initial scan; comparing, by the configured circuitry, the initial scan to a confidence threshold; and performing a subsequent scan based on the comparison. In some embodiments, capturing the image further comprises capturing, by the configured circuitry, a plurality of individual stack views of the image; saving, by the configured circuitry, each individual stack view; and executing, by the configured circuitry, a continuum diffusion process to fuse individual stacks together. In some embodiments, the plurality of individual stack views correspond to a plurality of distinct focal points. In some embodiments, a method may further include enhancing, by the configured circuitry, at least a viewability characteristic of the image using a machine-learning process. In another aspect, an apparatus for visualizing digitized slides may include at least a processor; and a memory communicatively connected to the at least processor, the memory containing instructions configuring the at least processor to retrieve a digitized slide; determine one or more visualization components of the digitized slide; generate a virtual slide corresponding to the digitized slide based on the one or more visualization components; and display a visualization of the virtual slide. In some embodiments, the memory contains instructions configuring the at least processor to determine, based on metadata associated with the digitized slide, that the digitized slide is a member of a set of digitized slides associated with at least one of a patient case or a tissue block. In some embodiments, displaying the virtual slide comprises displaying a plurality of virtual slides, including the virtual slide, corresponding to the set of digitized slides. In some embodiments, the one or more visualization components include at least one of a tissue section, an artifact, or an annotation. In some embodiments, the memory contains instructions configuring the at least processor to determine one or more user- configurable options associated with the virtual slide based on the one or more visualization components. In some embodiments, the one or more user-configurable options are determined by accessing a look-up table indexed by the one or more visualization components. In some embodiments, the visualization is displayed via the whole slide image viewer, and wherein the one or more user-configurable options are presented to a user via a user interface of a whole slide image viewer. In some embodiments, the memory contains instructions configuring the at least processor to receive a request to customize the visualization. In some embodiments, the memory contains instructions configuring the at least processor to receive a request to display a second 8 Attorney Docket No.1519-029PCT1 visualization of a different virtual slide. In some embodiments, the memory contains instructions configuring the at least processor to determine a recommended set of visualization components to include in the visualization; and determine a revised set of visualization components to include in the visualization based on a user selection. In some embodiments, the memory contains instructions configuring the at least processor to determine that the digitized slide corresponds to an intra-serial section slide based on a presence of a plurality of serial sections in the digitized slide; classify the plurality of serial sections into a reference serial section and one or more remaining serial sections; and align the one or more remaining serial sections to the reference serial section, yielding a plurality of aligned serial sections, wherein the visualization of the virtual slide includes the plurality of aligned serial sections. In some embodiments, the one or more remaining serial sections are aligned with the reference serial section by computing, independently for each of the one or more remaining serial sections, one or more registration transforms relative to the reference serial section. In some embodiments, the one or more registration transforms are computed based on a macro image of the digitized slide, the macro image being acquired using a macro camera and having a field of view that covers each of the plurality of serial sections. In some embodiments, the memory contains instructions configuring the at least processor to store the one or more registration transforms in a non-volatile storage medium; acquire a whole slide image (WSI) having a higher magnification than the macro image; compute based on the one or more stored registration transforms, one or more corresponding high-magnification registration transforms applicable to the WSI; apply the one or more high magnification registration transforms to the plurality of serial sections within the WSI to yield a virtual WSI having a plurality of aligned serial sections, wherein displaying the visualization of the virtual slide comprises displaying a visualization of the virtual WSI. In some embodiments, the plurality of aligned serial sections are displayed in the same order that the corresponding plurality of serial sections appear on the digital slide. In some embodiments, the plurality of aligned serial sections are spatially arranged within the visualization based on a user- selected configuration. In some embodiments, the plurality of aligned serial sections are spatially arranged in a compact representation such that the plurality of aligned serial sections appear closer to one another in the visualization than in the digitized slide. In some embodiments, the one or more visualization components include at least one annotation, wherein the at least one annotation is included in the visualization based on a user-configurable filter, and wherein 9 Attorney Docket No.1519-029PCT1 aligning the one or more remaining serial sections to the reference serial section includes aligning the at least one annotation to the reference serial section. In another aspect, a method for visualizing digitized slides may include retrieving, by at least a processor, a digitized slide; determining, by the at least a computer processor, at least a visualization component of the digitized slide; generating, by the at least a computer processor, a virtual slide corresponding to the digitized slide based on the at least a visualization component, displaying, by the at least a computer processor and at least a display, a visualization of the virtual slide. In some embodiments, a method may further include determining, by the at least a computer processor, based on metadata associated with the digitized slide, that the digitized slide is a member of a set of digitized slides associated with at least one of a patient case or a tissue block. In some embodiments, displaying the virtual slide comprises displaying a plurality of virtual slides, including the virtual slide, corresponding to the set of digitized slides. In some embodiments, the at least a visualization component include at least one of a tissue section, an artifact, or an annotation. In some embodiments, a method may further include determining, by the at least a computer processor, at least a user-configurable option associated with the virtual slide based on the at least a visualization component. In some embodiments, the at least a user- configurable option is determined by accessing a look-up table indexed by the at least a visualization component. In some embodiments, the visualization is displayed via a whole slide image viewer, and wherein the at least a user-configurable option is presented to a user via a user interface of the whole slide image viewer. In some embodiments, a method may further include receiving, by the at least a computer processor, a request to customize the visualization. In some embodiments, a method may further include receiving, by the at least a computer processor, a request to display a second visualization of a different virtual slide. In some embodiments, a method may further include determining, by the at least a computer processor, a recommended set of visualization components to include in the visualization; and determining, by the at least a computer processor, a revised set of visualization components to include in the visualization based on a user selection. In some embodiments, a method may further include determining, by the at least a computer processor, that the digitized slide corresponds to an intra-serial section slide based on a presence of a plurality of serial sections in the digitized slide; classifying, by the at least a computer processor, the plurality of serial sections into a reference serial section and at least a remaining serial section; and aligning, by the at least a computer processor, the at least a 10 Attorney Docket No.1519-029PCT1 remaining serial section to the reference serial section, yielding a plurality of aligned serial sections, wherein the visualization of the virtual slide includes the plurality of aligned serial sections. In some embodiments, the at least a remaining serial section is aligned with the reference serial section by computing, independently for each of the at least a remaining serial section, at least a registration transform relative to the reference serial section. In some embodiments, the at least a registration transform is computed based on a macro image of the digitized slide, the macro image being acquired using a macro camera and having a field of view that covers each of the plurality of serial sections. In some embodiments, a method may further include storing, by the at least a computer processor, the at least a registration transform in a non-volatile storage medium; acquiring, by the at least a computer processor, a whole slide image (WSI) having a higher magnification than the macro image; computing, by the at least a computer processor, based on the at least a stored registration transform, at least a corresponding high-magnification registration transform applicable to the WSI; applying, by the at least a computer processor, the at least a high magnification registration transform to the plurality of serial sections within the WSI to yield a virtual WSI having a plurality of aligned serial sections wherein displaying the visualization of the virtual slide comprises displaying a visualization of the virtual WSI. In some embodiments, the plurality of aligned serial sections are displayed in the same order that the corresponding plurality of serial sections appear on the digital slide. In some embodiments, the plurality of aligned serial sections are spatially arranged within the visualization based on a user-selected configuration. In some embodiments, the plurality of aligned serial sections are spatially arranged in a compact representation such that the plurality of aligned serial sections appear closer to one another in the visualization than in the digitized slide. In some embodiments, the at least a visualization component includes at least one annotation, wherein the at least one annotation is included in the visualization based on a user-configurable filter, and wherein aligning the at least a remaining serial section to the reference serial section includes aligning the at least one annotation to the reference serial section. In another aspect, an apparatus for imaging a slide may include at least an optical system, including an optical sensor; a slide port configured to hold a slide; at least a processor; and a memory communicatively connected to the at least processor, the memory containing instructions configuring the at least processor to receive at least a region of interest; capture, using the at least an optical system, a first image of the slide at a first position within the at least 11 Attorney Docket No.1519-029PCT1 a region of interest; identify a focus pattern as a function of the first image and the first position; extrapolate a focus distance for a second position as a function of the focus pattern; and capture, using the at least an optical system, a second image of the slide at a second position and at the focus distance. In some embodiments, an apparatus may further include an actuator mechanism mechanically connected to a mobile element; and wherein the memory contains instructions configuring the at least processor to, using the actuator mechanism, move the mobile element into the second position. In some embodiments, identifying the focus pattern comprises identifying a row; capturing a plurality of first images at the first location, wherein each of the plurality of first images has a different focus distance; determining an optimally focused first image of the plurality of first images having an optimal focus; and identifying the focus pattern using a focus distance of a plurality of optimally focused images at a set of points along the row. In some embodiments, the row further includes the second position and the instructions further configure the processor to capture, using the optical system, a plurality of second images at the second location, wherein each of the plurality of second images have a different focus distance; determine an optimally focused second image of the plurality of second images having an optimal focus; identify the focus pattern using the focus distance of the optimally focused first image and the optimally focused second image; and extrapolate a third focus distance for a third position as a function of the focus pattern. In some embodiments, the third position is located outside of the row. In some embodiments, the third position is located within a different region of interest than the first position. In some embodiments, identifying the row comprises identifying the row based on a first row sample presence score from a first set of row sample presence scores. In some embodiments, identifying the row based on the first row sample presence score comprises determining the row whose adjacent rows have the highest sample presence scores from a second set of sample presence scores, wherein the second set of sample presence scores is determined using machine vision. In some embodiments, the instructions further configure the processor to identify a point within the row that has a maximum point sample presence score, using machine vision. In some embodiments, identifying the plane comprises identifying a plurality of points and a plurality of optimal focuses at the plurality of points; and generating the plane as a function of a subset of the plurality of points and a corresponding subset of optimal focuses at those points. In some embodiments, identifying the focus pattern further comprises updating the focus pattern, wherein updating the focus pattern 12 Attorney Docket No.1519-029PCT1 comprises identifying an additional point and an optimal focus at the additional point; and updating the focus pattern as a function of the additional point and optimal focus at the additional point. In some embodiments, capturing the second image comprises capturing a plurality of images taken with focus distance based on the focus pattern; and constructing the second image from the plurality of images. In some embodiments, the memory contains instructions configuring the at least processor to capture, using the optical system, a low magnification image of the slide, wherein the low magnification image has a magnification lower than that of the first image; identify, using machine vision, the at least a region of interest within the low magnification image; and determine if either of the first image or the second image contains a sample. In another aspect, a method of imaging a slide may include using at least a processor, receiving at least a region of interest; using at least a processor and at least an optical system, capturing a first image of the slide at a first position within the at least a region of interest; using at least a processor, identifying a focus pattern as a function of the first image and the first position; using at least a processor, extrapolating a focus distance for a second position as a function of the focus pattern; and using at least a processor and the at least an optical system, capturing a second image of the slide at a second position and at the focus distance. In some embodiments, a method may further include using an actuator mechanism, moving a mobile element into the second position. In some embodiments, identifying the focus pattern comprises identifying a row which includes the first position; capturing a plurality of first images at the first location, wherein each of the plurality of first images has a different focus distance; determining an optimally focused first image of the plurality of first images having an optimal focus; and identifying the focus pattern using a focus distance of a plurality of optimally focused images at a set of points along the row. In some embodiments, the row further includes the second position and the method further comprises using at least a processor and the optical system, capturing a plurality of second images at the second location, wherein each of the plurality of second images have a different focus distance; using at least a processor, determining an optimally focused second image of the plurality of second images having an optimal focus; using at least a processor, identifying the focus pattern using the focus distance of the optimally focused first image and the optimally focused second image; and using at least a processor, extrapolating a third focus distance for a third position as a function of the focus pattern. In some embodiments, 13 Attorney Docket No.1519-029PCT1 the third position is located outside of the row. In some embodiments, the third position is located within a different region of interest than the first position. In some embodiments, identifying the row comprises identifying the row based on a first row sample presence score from a first set of row sample presence scores. In some embodiments, identifying the row based on the first row sample presence score comprises determining the row whose adjacent rows have the highest sample presence scores from a second set of sample presence scores, wherein the second set of sample presence scores is determined using machine vision. In some embodiments, a method may further include identifying a point within the row that has a maximum point sample presence score, using machine vision. In some embodiments, identifying the plane comprises identifying a plurality of points and a plurality of optimal focuses at the plurality of points; and generating the plane as a function of a subset of the plurality of points and a corresponding subset of optimal focuses at those points. In some embodiments, identifying the focus pattern further comprises updating the focus pattern, wherein updating the focus pattern comprises identifying an additional point and an optimal focus at the additional point; and updating the focus pattern as a function of the additional point and optimal focus at the additional point. In some embodiments, capturing the second image comprises capturing a plurality of images taken with focus distance based on the focus pattern; and constructing the second image from the plurality of images. In some embodiments, a method may further include using at least a processor and the optical system, capturing a low magnification image of the slide, wherein the low magnification image has a magnification lower than that of the first image; using at least a processor and machine vision, identifying the at least a region of interest within the low magnification image; and using at least a processor, determining if either of the first image or the second image contains a sample. DESCRIPTION OF DRAWINGS For the purpose of illustrating the invention, the drawings show aspects of one or more embodiments of the invention. However, it should be understood that the present invention is not limited to the precise arrangements and instrumentalities shown in the drawings, wherein: FIG.1 is a schematic diagram of an exemplary embodiment of a system for inline quality control of slide digitization; FIG.2 is a flow diagram of an exemplary embodiment of a data flow for inline quality control; 14 Attorney Docket No.1519-029PCT1 FIGS.3A-3B are diagrams of an exemplary embodiment of a localization module of a data flow for inline quality control; FIG.4 is a diagram of an exemplary embodiment of a biopsy plane estimation module of a data flow for inline quality control; FIG.5 is a diagram of an exemplary embodiment of a focus sampling module of a data flow for inline quality control; FIGS.6A-6B are diagrams of an exemplary embodiment of a z-stack acquisition module of a data flow for inline quality control; FIGS.7A-7B are diagrams of an exemplary embodiment of a stitching module of a data flow for inline quality control; FIG.8 is a flow diagram of an exemplary embodiment of a method for inline quality control of slide digitization; FIG.9 is a diagram depicting an exemplary embodiment of a system for digitizing a slide; FIG.10 is a flow diagram depicting an exemplary embodiment of a method of digitizing a slide; FIG.11 is a simplified diagram illustrating an exemplary system for slide digitization, in accordance with some embodiments; FIG.12 is a simplified diagram illustrating different example serial sections, in accordance with some embodiments; FIG.13 is a flow chart illustrating an exemplary method for slide digitization, in accordance with some embodiments; FIG.14 is a simplified diagram illustrating serial sections on different slides at various stages during slide digitization, in accordance with some embodiments; FIG.15 is a simplified diagram illustrating serial sections on the same slide at various stages during slide digitization, in accordance with some embodiments; FIG.16 is a flow chart illustrating an example method for slide digitization, in accordance with some embodiments. FIG.17 is a block diagram of an exemplary embodiment of an apparatus for inline scanned image enrichment; FIG.18 is a block diagram illustrating an exemplary embodiment of a method for visualization of digitized slides; 15 Attorney Docket No.1519-029PCT1 FIG.19 is a block diagram depicting an exemplary embodiment of an apparatus for slide imaging; FIG.20 is a box diagram of an exemplary machine learning model; FIG.21 is a diagram of an exemplary neural network; FIG.22 is a diagram of an exemplary neural network node; FIG.23 is a block diagram of a computing system that can be used to implement any one or more of the methodologies disclosed herein and any one or more portions thereof. The drawings are not necessarily to scale and may be illustrated by phantom lines, diagrammatic representations, and fragmentary views. In certain instances, details that are not necessary for an understanding of the embodiments or that render other details difficult to perceive may have been omitted. Like reference symbols in the various drawings indicate like elements. DETAILED DESCRIPTION Over the last several years, an increasing amount of clinical patient data has become digitized, and progress has been made in building systems that process clinical patient data in an electronic format. For example, cloud-based systems can collect digitized health records from across institutions (e.g., hospitals, clinics, etc.) and can analyze patient data at a scale that was not attainable until recently. For example, de-identification of electronic health records has opened up the possibility of learning from de-identified data within and across medical institutions/facilities that are owners of the de-identified data. Variations of federated learning approaches are now being used to facilitate such learning within and across data repositories. As the amount of data and processing scale increases, there is an emerging need for techniques that can automate the process of acquiring clinically relevant digital data. Digitized slides, such as whole slide images (WSIs), are one such type of digital data. Clinical and research institutions use slides containing biomedical specimens (e.g., tissue samples) for a wide range of applications such as disease diagnosis. The ability to digitize slides at scale may facilitate a variety of downstream applications such as training machine learning models and performing large-scale data analysis. However, this approach to slide digitization may be manually intensive, involving intervention by a clinician at various steps of the digitization process. Moreover, this approach may be error-prone or inefficient. For example, as further explained below, portions of the output 16 Attorney Docket No.1519-029PCT1 image may be out of focus and may contain unwanted artifacts, such as ghosting and banding. As another example, the specimen of interest may be missing from the portion of the slide that is scanned. To prevent such digitization errors from interfering with downstream applications, it may be possible to perform a post-imaging quality check to determine whether a slide image meets a given acceptance criteria. However, post-imaging quality checks offer limited remedial options. For example, one remedial option may be to remove an image of a slide that does not meet the acceptance criteria from the data set, but this undesirably results in a smaller set of data for downstream applications, and the resources that were used to prepare and scan the slide would be wasted. Another remedial option may be to rescan a slide that does not meet the acceptance criteria, but this approach may be inefficient – e.g., this approach may involve rescanning an entire slide even when the error impacts a portion of the slide, the slide may have already been physically removed from the scanner or placed into storage, or the like. These inefficiencies may result in wasted resources, delayed diagnoses, increased manual intervention in the digitization process, low throughput, or the like. For example, a given slide may undergo multiple passes through the scanner before an image that passes the quality checks is produced, resulting in additional time consumption to set up and wind down each pass through the scanner. Ultimately, digitization errors may have the effect of significantly reducing the scanning throughput and thereby limiting the scalability of the slide digitization process. Therefore, an opportunity exists for slide digitization with inline quality controls – e.g., quality controls that are performed concurrently with the steps of the slide digitization process and/or prior to removing the slide from the scanner. Inline quality controls may facilitate increased automation, quality, and throughput in the slide digitization process while mitigating the downsides associated with inadequate quality control, e.g., the post-imaging quality control described above. For example, inline quality controls may facilitate throughput-enhancing remedial options, such as rescanning selected portions of a slide determined to be defective (rather than an entire slide) and initiating rescans prior to a slide being removed from the scanner. Illustratively, inline quality controls may enable scanning a slide, performing quality checks, and rescanning areas identified as defective in a single pass through the scanner, which in turn increases the throughput of high-quality scanned images. In some embodiments, a system and/or method described herein may be used for high volume scanning. 17 Attorney Docket No.1519-029PCT1 FIG.1 is a simplified diagram of a system 100 for inline quality control of slide digitization according to some embodiments. System 100 may include a processor. Processor may include, without limitation, any processor described in this disclosure. Processor may be included in a computing device. System 100 may include at least a processor and a memory communicatively connected to the at least a processor, the memory containing instructions configuring the at least a processor to perform one or more processes described herein. Computing device may include any computing device as described in this disclosure, including without limitation a microcontroller, microprocessor, digital signal processor (DSP) and/or system on a chip (SoC) as described in this disclosure. Computing device may include, be included in, and/or communicate with a mobile device such as a mobile telephone or smartphone. Computing device may include a single computing device operating independently, or may include two or more computing device operating in concert, in parallel, sequentially or the like; two or more computing devices may be included together in a single computing device or in two or more computing devices. Computing device may interface or communicate with one or more additional devices as described below in further detail via a network interface device. Network interface device may be utilized for connecting computing device to one or more of a variety of networks, and one or more devices. Examples of a network interface device include, but are not limited to, a network interface card (e.g., a mobile network interface card, a LAN card), a modem, and any combination thereof. Examples of a network include, but are not limited to, a wide area network (e.g., the Internet, an enterprise network), a local area network (e.g., a network associated with an office, a building, a campus or other relatively small geographic space), a telephone network, a data network associated with a telephone/voice provider (e.g., a mobile communications provider data and/or voice network), a direct connection between two computing devices, and any combinations thereof. A network may employ a wired and/or a wireless mode of communication. In general, any network topology may be used. Information (e.g., data, software etc.) may be communicated to and/or from a computer and/or a computing device. Computing device may include but is not limited to, for example, a computing device or cluster of computing devices in a first location and a second computing device or cluster of computing devices in a second location. Computing device may include one or more computing devices dedicated to data storage, security, distribution of traffic for load balancing, and the like. Computing device may distribute one or more computing tasks as 18 Attorney Docket No.1519-029PCT1 described below across a plurality of computing devices of computing device, which may operate in parallel, in series, redundantly, or in any other manner used for distribution of tasks or memory between computing devices. Computing device may be implemented, as a non-limiting example, using a “shared nothing” architecture. Still referring to FIG.1, computing device may be designed and/or configured to perform any method, method step, or sequence of method steps in any embodiment described in this disclosure, in any order and with any degree of repetition. For instance, computing device may be configured to perform a single step or sequence repeatedly until a desired or commanded outcome is achieved; repetition of a step or a sequence of steps may be performed iteratively and/or recursively using outputs of previous repetitions as inputs to subsequent repetitions, aggregating inputs and/or outputs of repetitions to produce an aggregate result, reduction or decrement of one or more variables such as global variables, and/or division of a larger processing task into a set of iteratively addressed smaller processing tasks. Computing device may perform any step or sequence of steps as described in this disclosure in parallel, such as simultaneously and/or substantially simultaneously performing a step two or more times using two or more parallel threads, processor cores, or the like; division of tasks between parallel threads and/or processes may be performed according to any protocol suitable for division of tasks between iterations. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which steps, sequences of steps, processing tasks, and/or data may be subdivided, shared, or otherwise dealt with using iteration, recursion, and/or parallel processing. Still referring to FIG.1, as used in this disclosure, “communicatively connected” means connected by way of a connection, attachment or linkage between two or more relata which allows for reception and/or transmittance of information therebetween. For example, and without limitation, this connection may be wired or wireless, direct or indirect, and between two or more components, circuits, devices, systems, and the like, which allows for reception and/or transmittance of data and/or signal(s) therebetween. Data and/or signals therebetween may include, without limitation, electrical, electromagnetic, magnetic, video, audio, radio and microwave data and/or signals, combinations thereof, and the like, among others. A communicative connection may be achieved, for example and without limitation, through wired or wireless electronic, digital or analog, communication, either directly or by way of one or more 19 Attorney Docket No.1519-029PCT1 intervening devices or components. Further, communicative connection may include electrically coupling or connecting at least an output of one device, component, or circuit to at least an input of another device, component, or circuit. For example, and without limitation, via a bus or other facility for intercommunication between elements of a computing device. Communicative connecting may also include indirect connections via, for example and without limitation, wireless connection, radio communication, low power wide area network, optical communication, magnetic, capacitive, or optical coupling, and the like. In some instances, the terminology “communicatively coupled” may be used in place of communicatively connected in this disclosure. Still referring to FIG.1, in some embodiments, system 100 may be used to generate an image of slide and/or a sample on slide. As used herein, a “slide” is a container or surface holding a sample of interest. In some embodiments, slide may include a glass slide. In some embodiments, slide may include a formalin fixed paraffin embedded slide. In some embodiments, a sample on slide may be stained. In some embodiments, slide may be substantially transparent. In some embodiments, slide may include a thin, flat, and substantially transparent glass slide. In some embodiments, a transparent cover may be applied to slide such that a sample is between slide and this cover. A sample may include, in non-limiting examples, a blood smear, pap smear, body fluids, and non-biologic samples. In some embodiments, a sample on slide may include tissue. In some embodiments, sample on slide may be frozen. Still referring to FIG.1, in some embodiments, slide and/or a sample on slide may be illuminated. In some embodiments, system 100 may include a light source. As used herein, a “light source” is any device configured to emit electromagnetic radiation. In some embodiments, light source may emit a light having substantially one wavelength. In some embodiments, light source may emit a light having a wavelength range. Light source may emit, without limitation, ultraviolet light, visible light, and/or infrared light. In non-limiting examples, light source may include a light-emitting diode (LED), an organic LED (OLED) and/or any other light emitter. Such a light source may be configured to illuminate slide and/or sample on slide. In a non- limiting example, light source may illuminate slide and/or sample on slide from below. Still referring to FIG.1, in some embodiments, system 100 may include at least an optical system. As used in this disclosure, an "optical system" is an arrangement of one or more components which together act upon or employ electromagnetic radiation. In non-limiting 20 Attorney Docket No.1519-029PCT1 examples, electromagnetic radiation may include light, such as visible light, infrared light, UV light, and the like. An optical system may include one or more optical elements, including without limitation lenses, mirrors, windows, filters, and the like. An optical system may form an optical image that corresponds to an optical object. For instance, an optical system may form an optical image at or upon an optical sensor, which can capture, e.g., digitize, the optical image. In some cases, optical system may have at least a magnification. For instance, optical system may include an objective (e.g., microscope objective) and one or more reimaging optical elements that together produce an optical magnification. In some cases, optical magnification may be referred to herein as zoom. As used herein, an “optical sensor” is a device that measures light and converts the measured light into one or more signals; one or more signals may include, without limitation, one or more electrical signals. In some embodiments, optical sensor may include at least a photodetector. As used herein, a “photodetector” is a device that is sensitive to light and thereby able to detect light. In some embodiments, a photodetector may include a photodiode, a photoresistor, a photosensor, a photovoltaic chip, and the like. In some embodiments, optical sensor may include a plurality of photodetectors. Optical sensor may include, without limitation, a camera. Optical sensor may be in electronic communication with at least a processor of system 100. As used herein, “electronic communication” as used in this disclosure is a shared data connection between two or more devices. In some embodiments, system 100 may include two or more optical sensors. In some embodiments, a scanner may include at least an optical system. For example, scanner 101, scanner 102, and/or scanner 109 may include at least an optical system. Still referring to FIG.1, in some embodiments, optical system may include a camera. In some cases, a camera may include one or more optics. Exemplary non-limiting optics include spherical lenses, aspherical lenses, reflectors, polarizers, filters, windows, aperture stops, and the like. In some embodiments, one or more optics associated with a camera may be adjusted in order to, in non-limiting examples, change the zoom, depth of field, and/or focus distance of the camera. In some embodiments, one or more of such settings may be configured to detect a feature of a sample on slide. In some embodiments, one or more of such settings may be configured based on a scanning parameter, as described herein. In some embodiments, camera may capture images at a low depth of field. In a non-limiting example, camera may capture images such that a first depth of sample is in focus and a second depth of sample is out of focus. 21 Attorney Docket No.1519-029PCT1 In some embodiments, an autofocus mechanism may be used to determine focus distance. In some embodiments, focus distance may be set by parameter set. In some embodiments, camera may be configured to capture a plurality of images at different focus distances. In a non-limiting example, camera may capture a plurality of images at different focus distances, such that images are captured where each focus depth of the sample is in focus in at least one image. In some embodiments, at least a camera may include an image sensor. Exemplary non-limiting image sensors include digital image sensors, such as without limitation charge-coupled device (CCD) sensors and complimentary metal-oxide-semiconductor (CMOS) sensors. In some embodiments, a camera may be sensitive within a non-visible range of electromagnetic radiation, such as without limitation infrared. Still referring to FIG.1, as used herein, “image data” is information representing at least a physical scene, space, and/or object. Image data may include, for example, information representing a sample, slide, or region of a sample or slide. In some cases, image data may be generated by a camera. “Image data” may be used interchangeably through this disclosure with “image,” where image is used as a noun. An image may be optical, such as without limitation where at least an optic is used to generate an image of an object. An image may be digital, such as without limitation when represented as a bitmap. Alternatively, an image may include any media capable of representing a physical scene, space, and/or object. Alternatively, where “image” is used as a verb, in this disclosure, it refers to generation and/or formation of an image. Still referring to FIG.1, in some embodiments, system 100 may include a slide port. In some embodiments, slide port may be configured to hold slide. In some embodiments, slide port may include one or more alignment features. As used herein, an “alignment feature” is a physical feature that helps to secure a slide in place and/or align a slide with another component of an apparatus. In some embodiments, alignment feature may include a component which keeps slide secure, such as a clamp, latch, clip, recessed area, or another fastener. In some embodiments, slide port may allow for easy removal or insertion of slide. In some embodiments, slide port may include a transparent surface through which light may travel. In some embodiments, slide may rest on and/or may be illuminated by light traveling through such a transparent surface. In some embodiments, slide port may be mechanically connected to an actuator mechanism as described below. In some embodiments, a scanner may include a slide port. For example, scanner 101, scanner 102, and/or scanner 109 may include a slide port. 22 Attorney Docket No.1519-029PCT1 Still referring to FIG.1, in some embodiments, system 100 may include an actuator mechanism. As used herein, an “actuator mechanism” is a mechanical component configured to change the relative position of a slide and an optical system. In some embodiments, actuator mechanism may be mechanically connected to slide, such as slide in slide port. In some embodiments, actuator mechanism may be mechanically connected to slide port. For example, actuator mechanism may move slide port in order to move slide. In some embodiments, actuator mechanism may be mechanically connected to at least an optical system. In some embodiments, actuator mechanism may be mechanically connected to a mobile element. A mobile element may include a movable or portable object, component, and/or device within system 100 such as, without limitation, a slide, a slide port, or an optical system. In some embodiments, a mobile element may move such that optical system is positioned correctly with respect to slide such that optical system may capture an image of slide according to a scanning parameter. In some embodiments, actuator mechanism may be mechanically connected to an item selected from the list consisting of slide port, slide, and at least an optical system. In some embodiments, actuator mechanism may be configured to change the relative position of slide and optical system by moving slide port, slide, and/or optical system. In some embodiments, a scanner may include an actuator mechanism. For example, scanner 101, scanner 102, and/or scanner 109 may include an actuator mechanism. Still referring to FIG.1, actuator mechanism may include a component of a machine that is responsible for moving and/or controlling a mechanism or system. Actuator mechanism may, in some embodiments, require a control signal and/or a source of energy or power. In some cases, a control signal may be relatively low energy. Exemplary control signal forms include electric potential or current, pneumatic pressure or flow, or hydraulic fluid pressure or flow, mechanical force/torque or velocity, or even human power. In some cases, an actuator may have an energy or power source other than control signal. This may include a main energy source, which may include for example electric power, hydraulic power, pneumatic power, mechanical power, and the like. In some embodiments, upon receiving a control signal, actuator mechanism responds by converting source power into mechanical motion. In some cases, actuator mechanism may be understood as a form of automation or automatic control. Still referring to FIG.1, in some embodiments, actuator mechanism may include a hydraulic actuator. A hydraulic actuator may consist of a cylinder or fluid motor that uses 23 Attorney Docket No.1519-029PCT1 hydraulic power to facilitate mechanical operation. Output of hydraulic actuator mechanism may include mechanical motion, such as without limitation linear, rotatory, or oscillatory motion. In some embodiments, hydraulic actuator may employ a liquid hydraulic fluid. As liquids, in some cases, are incompressible, a hydraulic actuator can exert large forces. Additionally, as force is equal to pressure multiplied by area, hydraulic actuators may act as force transformers with changes in area (e.g., cross sectional area of cylinder and/or piston). An exemplary hydraulic cylinder may consist of a hollow cylindrical tube within which a piston can slide. In some cases, a hydraulic cylinder may be considered single acting. Single acting may be used when fluid pressure is applied substantially to just one side of a piston. Consequently, a single acting piston can move in only one direction. In some cases, a spring may be used to give a single acting piston a return stroke. In some cases, a hydraulic cylinder may be double acting. Double acting may be used when pressure is applied substantially on each side of a piston; any difference in resultant force between the two sides of the piston causes the piston to move. Still referring to FIG.1, in some embodiments, actuator mechanism may include a pneumatic actuator mechanism. In some cases, a pneumatic actuator may enable considerable forces to be produced from relatively small changes in gas pressure. In some cases, a pneumatic actuator may respond more quickly than other types of actuators, for example hydraulic actuators. A pneumatic actuator may use compressible fluid (e.g., air). In some cases, a pneumatic actuator may operate on compressed air. Operation of hydraulic and/or pneumatic actuators may include control of one or more valves, circuits, fluid pumps, and/or fluid manifolds. Still referring to FIG.1, in some cases, actuator mechanism may include an electric actuator. Electric actuator mechanism may include any of electromechanical actuators, linear motors, and the like. In some cases, actuator mechanism may include an electromechanical actuator. An electromechanical actuator may convert a rotational force of an electric rotary motor into a linear movement to generate a linear movement through a mechanism. Exemplary mechanisms, include rotational to translational motion transformers, such as without limitation a belt, a screw, a crank, a cam, a linkage, a scotch yoke, and the like. In some cases, control of an electromechanical actuator may include control of electric motor, for instance a control signal may control one or more electric motor parameters to control electromechanical actuator. Exemplary non-limitation electric motor parameters include rotational position, input torque, 24 Attorney Docket No.1519-029PCT1 velocity, current, and potential. Electric actuator mechanism may include a linear motor. Linear motors may differ from electromechanical actuators, as power from linear motors is output directly as translational motion, rather than output as rotational motion and converted to translational motion. In some cases, a linear motor may cause lower friction losses than other devices. Linear motors may be further specified into at least 3 different categories, including flat linear motor, U-channel linear motors and tubular linear motors. Linear motors may be directly controlled by a control signal for controlling one or more linear motor parameters. Exemplary linear motor parameters include without limitation position, force, velocity, potential, and current. Still referring to FIG.1, in some embodiments, an actuator mechanism may include a mechanical actuator mechanism. In some cases, a mechanical actuator mechanism may function to execute movement by converting one kind of motion, such as rotary motion, into another kind, such as linear motion. An exemplary mechanical actuator includes a rack and pinion. In some cases, a mechanical power source, such as a power take off may serve as power source for a mechanical actuator. Mechanical actuators may employ any number of mechanisms, including for example without limitation gears, rails, pulleys, cables, linkages, and the like. Still referring to FIG.1, in some embodiments, actuator mechanism may be in electronic communication with actuator controls. As used herein, “actuator controls” is a system configured to operate actuator mechanism such that a slide and an optical system reach a desired relative position. In some embodiments, actuator controls may operate actuator mechanism based on input received from a user interface. In some embodiments, actuator controls may be configured to operate actuator mechanism such that optical system is in a position to capture an image of an entire sample. In some embodiments, actuator controls may be configured to operate actuator mechanism such that optical system is in a position to capture an image using settings of a particular scanning parameter. In some embodiments, actuator controls may be configured to operate actuator mechanism such that optical system is in a position to capture an image of a region of interest, a particular horizontal row, a particular point, a particular focus depth, and the like. Electronic communication between actuator mechanism and actuator controls may include transmission of signals. For example, actuator controls may generate physical movements of actuator mechanism in response to an input signal. In some embodiments, input signal may be received by actuator controls from processor or input interface. 25 Attorney Docket No.1519-029PCT1 Still referring to FIG.1, in some embodiments, system 100 may include a user interface. User interface may include output interface and input interface. In some embodiments, output interface may include one or more elements through which system 100 may communicate information to a user. In a non-limiting example, output interface may include a display. A display may include a high resolution display. A display may output images, videos, and the like to a user. In another non-limiting example, output interface may include a speaker. A speaker may output audio to a user. In another non-limiting example, output interface may include a haptic device. A haptic device may output haptic feedback to a user. Still referring to FIG.1, in some embodiments, input interface may include controls for operating system. Such controls may be operated by a user. Input interface may include, in non- limiting examples, a camera, microphone, keyboard, touch screen, mouse, joystick, foot pedal, button, dial, and the like. Input interface may accept, in non-limiting examples, mechanical input, audio input, visual input, text input, and the like. In some embodiments, input interface may approximate controls of a microscope. Still referring to FIG.1, system 100 may perform inline quality control of slide digitization. In a non-limiting embodiment, system 100 may include one or more scanners 101- 109 that are communicatively coupled to a device 110. Scanners 101-109 generally include devices or systems used to digitize slides containing biomedical specimens (e.g., tissue samples), such as digital cameras, digital microscopes, digital pathology scanners, or the like. Device 110 generally includes a computing device or system, such as personal computer, an on-premises server, a cloud-based server, or the like. In some embodiments, device 110 may be directly connected to and/or integrated within one or more of scanners 101-109. Additionally, or alternately, scanners 101-109 may be communicatively coupled to device 110 via a network 115. Network 115 can include one or more local area networks (LANs), wide area networks (WANs), wired networks, wireless networks, the Internet, or the like. Illustratively, scanners 101-109 may communicate with device 110 over network 115 using the TCP/IP protocol or other suitable networking protocols. During operation, scanners 101-109 capture digital images 121-129 of the slides and send them to device 110, e.g., via network 115. For efficient storage and/or transmission, images 121-129 may be compressed prior to or during transmission. Security measures such as encryption, authentication (including multi-factor authentication), SSL, HTTPS, and other security techniques may also be applied. 26 Attorney Docket No.1519-029PCT1 Still referring to FIG.1, in some embodiments, device 110 may be configured to control the operation of scanners 101-109 to automate the slide digitization process. For example, device 110 may send instructions to scanners 101-109 to scan or rescan selected portions of a slide (e.g., areas identified by x-, y-, and/or z-coordinates or bounding boxes). Device 110 may further select parameters associated with scanners 101-109, such as magnification level, focus settings, scanning pattern, or the like. Still referring to FIG.1, as discussed above, automating the slide digitization process using device 110 may present challenges. For example, one or more of images 121-129 may be out of focus or may capture a different portion of the slide than intended, e.g., the actual coordinates of the image may be offset from the intended coordinates. In embodiments where a series of images of a given slide are captured (e.g., when the slide is scanned using a grid scanning pattern), the process of stitching the images together to generate an aggregated image of the slide may introduce artifacts, such as ghosting or banding. These defects may interfere with downstream applications that receive the digitized slide images output by device 110. For example, the defects may result in a reduction in performance of a machine learning model trained using the digitized slides due to the reduced size and/or quality of the training data set. Moreover, the defects may be identified too late to efficiently take remedial action, e.g., a defective image may be removed from the data set (resulting in wasted resources that went into scanning the image), or the slide may be placed back into the scanner for another scanning pass (resulting in additional setup and takedown time). Such remedial actions may involve manual intervention, such that the level of automation in the slide digitization process is reduced. Still referring to FIG.1, therefore, according to some embodiments, device 110 includes an inline quality control program 150, which may address one or more of the challenges identified above. Namely, as depicted in FIG.1, device 110 includes a processor 130 (e.g., one or more hardware processors) coupled to a memory 140 (e.g., one or more non-transitory memories). Memory 140 stores instructions and/or data corresponding to inline quality control program 150. When executed by processor 130, inline quality control program 150 causes processor 130 to perform operations associated with inline quality control of slide digitization based on images 121-129. Illustrative embodiments of data flows implemented by inline quality control program 150 are described in further detail below with reference to FIG.2. 27 Attorney Docket No.1519-029PCT1 Still referring to FIG.1, during execution of inline quality control program 150, processor 130 may execute one or more neural network models, such as neural network model 160. Neural network model 160 is trained to make predictions (e.g., inferences) based on input data. Neural network model 160 includes a configuration 162, which defines a plurality of layers of neural network model 160 and the relationships among the layers. Illustrative examples of layers include input layers, output layers, convolutional layers, densely connected layers, merge layers, and the like. In some embodiments, neural network model 160 may be configured as a deep neural network with at least one hidden layer between the input and output layers. Connections between layers can include feed-forward connections or recurrent connections. Still referring to FIG.1, one or more layers of neural network model 160 is associated with trained model parameters 164. The trained model parameters 164 include a set of parameters (e.g., weight and bias parameters of artificial neurons) that are learned according to a machine learning process. During the machine learning process, labeled training data is provided as an input to neural network model 160, and the values of trained model parameters 164 are iteratively adjusted until the predictions generated by neural network model 160 match the corresponding labels with a desired level of accuracy. Still referring to FIG.1, for improved performance, processor 130 may execute neural network model 160 using a graphical processing unit, a tensor processing unit, an application- specific integrated circuit, or the like. FIGS.2-7B are simplified diagrams of a data flow 200 for inline quality control according to some embodiments. FIG.2 provides an overview of data flow 200, and FIGS.3-7B illustrate certain modules of data flow 200 in greater detail. In some embodiments consistent with FIG.1, data flow 200 may be implemented using inline quality control program 150 in conjunction with other components and/or features of system 100, as further described below. Referring now to FIG.2, in general, data flow 200 includes a set of modules 210-260 (described in further detail below) that correspond to steps in the slide digitization process. Within each module, inline quality control methods are in place to trap errors that can be fixed at that stage below an acceptable threshold for downstream processing. Additionally, or alternately, the inline quality control methods associated with each module may accumulate deviations of observed features from acceptable thresholds and pass the deviations to the next module. In this sense, data flow 200 conceptually resembles a directed computation graph with directed edges 28 Attorney Docket No.1519-029PCT1 leading back to the beginning with information to perform a second pass selectively to fix errors. Most nodes in the graph have automatically determined quality acceptance tests baked in, with tuned thresholds that determine if the processed image proceeds to the next stage or is fed back to the beginning to correct for specific errors detected. For images that are finally output, the errors are within acceptable thresholds. Additionally, the errors that are present in those slides are demarcated for downstream algorithms to avoid or fix. For instance, generative models like diffusion models can leverage these delineated regions to perform inpainting which is then used for downstream tasks. Still referring to FIG.2, in some embodiments, system 100 performs a scan of a slide. System 100 may perform a scan of a slide by capturing at least a first image and capturing at least a second image. Performing a scan may include identifying a first scanning parameter. As used herein, a “scanning parameter” is a setting at which a device captures an image. In non- limiting examples, a scanning parameter may include a focal depth, a (x,y) position, a magnification level, an aperture, a shutter speed, an ISO sensitivity, and a level of backlighting. In some embodiments, a first scanning parameter may include which of a plurality of optical sensors to use to capture an image. For example, first scanning parameter may configure a device to capture a macro image of a whole slide and/or an entirety of a sample on a slide. As used herein, a “macro image” is an image in which entirety of a sample on a slide is in frame. In another example, at least a first image includes an image of a region of a sample. Such a macro image may be a wider angle image than subsequently captured images. Such a macro image may include a lower resolution image than a subsequently captured image. Such a macro image may be used to determine subsequent scanning parameters, as described herein. In some embodiments, a macro image may be captured at 1x magnification. In some embodiments, a macro image may be captured using a 5 megapixel camera. Still referring to FIG.2, in some embodiments, performing a scan of a slide may include, using an optical system, capturing at least a first image as a function of a first scanning parameter. In some embodiments, at least a first image may be captured at a first position as a function of first scanning parameter. In some embodiments, capturing at least a first image of a slide may include using actuator mechanism and/or actuator controls to move optical system and/or slide into desired positions. 29 Attorney Docket No.1519-029PCT1 Still referring to FIG.2, in some embodiments, system 100 may determine a first quality metric as a function of at least a first image. As used herein, a “quality metric” is a datum describing an assessment of a quality of an image, a quality of a section of an image, a quality of a feature of an image, or a combination thereof. In a non-limiting example, a quality metric may describe a degree to which an image and/or a section of an image is in focus. Non-limiting examples of quality metrics include localization quality metric, focus sampling quality metric, biopsy plane estimation quality metric, z-stack acquisition quality metric, and stitching quality metric, each of which is described further herein. Still referring to FIG.2, in some embodiments, system 100 may determine a second scanning parameter as a function of first quality metric. System 100 may, using an optical system, capture at least a second image as a function of second scanning parameter. Examples of quality metrics, and their associated scanning parameters are described further below. In some embodiments, at least a second image may be made up of a plurality of high resolution images. Such high resolution images may be stitched together to construct at least a second image which covers a desired area, such as an entire sample or region of interest. In some embodiments, at least a first image may be captured at a first location, and at least a second image may be captured at a second location. For example, a first location may be determined as a function of first scanning parameter, and may include a default location, a location input by a user, or the like. As an example, a second location may be determined as a function of a localization module as described further herein. Still referring to FIG.2, in some embodiments, system 100 may determine multiple quality metrics. For example, system 100 may determine localization quality metric, focus sampling quality metric, biopsy plane estimation quality metric, z-stack acquisition quality metric, and stitching quality metric simultaneously. In some embodiments, system 100 may determine one or more scanning parameters and/or capture one or more images based on such quality metrics. For example, system 100 may first determine localization quality metric, then may determine a scanning parameter and may capture a subsequent image using such scanning parameter. System 100 may then determine focus sampling quality metric, may determine a subsequent scanning parameter, and may capture a subsequent image using such scanning parameter. 30 Attorney Docket No.1519-029PCT1 Referring now to FIGS.2 and 3A-3B, first quality metric may include a localization quality metric; and determining second scanning parameter may include identifying at least a region of interest of slide 304. As used herein, a “region of interest” is a specific area within a slide, or a digital image of a slide, in which a feature is detected. A feature may include, in non- limiting examples, debris 308 sample 312, writing on a slide, a crack in a slide, a bubble, and the like. In some embodiments, a feature includes a sample, such as a biological sample. As used herein, a “localization quality metric” is a quality metric which assesses the performance of a localization module. In some embodiments, localization quality metric may include one or more composite measures used to evaluate the precision and reliability of localization module’s ability to detect and position the sample within an image. In other embodiments, localization quality metric may include one or more quantifiable measures utilized by localization module within the system designed to identify region of interest 316. As shown in FIGS.2 and 3A-3B, data flow 200 includes a localization module 210. As used herein, a “localization module” is an automatic system for identifying a region of interest. Localization module 210 receives an image 310 of slide 304 (shown in FIG.3A), such as one of images 121-129. In some embodiments, image 310 may be captured at a low magnification level (e.g., 1x) and the field of view may cover a substantial portion of slide 304 (e.g., the image may include most of slide 304, including regions containing the biomedical specimens). Still referring to FIGS.2 and 3A-3B, based on image 310, localization module 210 localizes regions of interest 316 on slide 304. For example, as illustrated by image 320, localization module 210 may identify portions of slide 304 that contain biomedical specimens and other slide attributes including but not limited to the color intensity of stained tissue (e.g., dark or faint), the size, shape, and position of the specimen, debris, cover slip boundaries, annotations (e.g., handwritten text, printed text, scanning codes, arrows and markings, or the like), or the like. Localizing regions of interest 316 on slide 304 may be performed by a variety of suitable image processing techniques, including techniques that use machine learning models trained to identify the selected regions of interest. Annotations may be identified using, for example, an optical character recognition (OCR) system. Still referring to FIGS.2 and 3A-3B, based on the identified areas of interest, localization module 210 may select one or more portions of the slide to be scanned at a higher magnification level (e.g., 10x, 20x, or 40x magnification). For example, localization module 210 may identify 31 Attorney Docket No.1519-029PCT1 one or more bounding boxes 332 that represent the portions of the slide to be scanned at the higher magnification level, e.g., portions containing the biomedical specimen. Localization module 210 may then form a grid 334 (or other suitable pattern) for each bounding box 332 that identifies the x, y, and/or z coordinates of the desired images to be captured at the higher resolution. In addition, localization module 210 may identify a scanning pattern 336 (e.g., a raster scanning pattern having a primary scanning direction and a secondary scanning direction) that specifies a particular sequence in which the desired images are to be acquired. In some embodiments, localization module 210 may identify one or more features during localization. Such features may include, in non-limiting examples, samples, bubbles, debris, and annotations. In some embodiments, localization module 210 may determine a localization module confidence score. As used herein, a “localization module confidence score” is a data structure describing a likelihood that a feature of a particular category is present in a particular location, region, or both. For example, a first localization module confidence score may be determined with respect to a particular region of a slide, where the first localization module confidence score describes the likelihood that the region contains an annotation. In this example, a second localization module confidence score may be determined with respect to a particular region of a slide, where the second localization module confidence score describes the likelihood that the region contains a biological sample. In some embodiments, localization module 210 may identify a region of interest as a function of one or more localization module confidence scores. For example, localization module 210 may identify a region of interest as a function of a localization module confidence score describing the likelihood that a biological sample is present. In some embodiments, high magnification probing may be used to determine whether a feature, such as a biological sample, is present. For example, high magnification images may be captured of a plurality of regions of a slide, and such high magnification images may be analyzed to determine whether a feature (such as a biological sample) is present. In some embodiments, high magnification probing may be used to determine a region of interest which does not include a region with a low probability of containing a feature such as a biological sample. Still referring to FIGS.2 and 3A-3B, in some embodiments, localization module 210 may select one or more reference points 338 within grid 334 for further imaging and analysis, such as biopsy plane estimation or focus sampling, as further described below. As used herein, a “reference point” is a point within a grid whose focal distance, determined by one or more 32 Attorney Docket No.1519-029PCT1 predetermined criteria, is used to calculate the focal distance of another point. In some embodiments, the optimal focal distance of a reference point is used to calculate the optimal focal distance of a different point. For example, reference point 338 may correspond to a point at the center of bounding box 332, a point having the maximum color intensity within bounding box 332, a point determined to be the most likely to contain a biomedical specimen (e.g., as determined using a machine learning model), or another suitable selection criteria. Still referring to FIGS.2 and 3A-3B, in some embodiments, localization module 210 may include one or more inline quality controls. For example, and without limitation, localization module 210 characterizes the content present on the slide, may check that image 310 is in focus, that image 310 includes the biological specimen within the field of view, that an amount of debris on the slide or other unwanted artifacts or noise in image 310 does not exceed a predetermined threshold, that a confidence level associated with identifying areas of interest exceeds a predetermined threshold, or the like. When one or more checks returns a negative result, one or more remedial actions may be taken, such as recapturing image 310 with different exposure settings, bringing image processing/ML algorithms to enhance the image etc., indicating to a lab admin the slide preparation, or the like. In some embodiments, the inline quality checks may output information associated with the results of the quality checks, such as a numerical score or an indication of areas of a slide that were found to be low quality, that can be used by subsequent modules in data flow 200. Still referring to FIGS.2 and 3A-3B, in some embodiments, the areas of interest identified by localization module 210 may be provided to downstream applications for further processing. For example, localization module 210 may provide a mask to delineate areas of interest with artifacts that should be avoided, inpainted, or otherwise handled by the downstream application. Referring now to FIGS.2 and 4, first quality metric may include biopsy plane estimation quality metric; and determining second scanning parameter may include identifying a plane within a region of interest which contains a biological specimen. As used herein, a “biopsy plane estimation quality metric” is a quality metric which assesses the performance of a biopsy plane estimation module. In some embodiments, biopsy plane estimation quality metric may include one or more composite measures used to evaluate the precision and reliability of biopsy plane estimation module’s ability to identify a plane within a region of interest which contains a 33 Attorney Docket No.1519-029PCT1 biological specimen. In other embodiments, localization quality metric may include one or more quantifiable measures utilized by localization module within the system designed to identify a plane within a region of interest which contains a biological specimen. As depicted in FIGS.2 and 4, data flow 200 may include a biopsy plane estimation module 220. As used herein, a “biopsy plane estimation module” is an automatic system for identifying a plane within a region of interest which contains a biological specimen. Biopsy plane estimation module 220 instructs a scanner 410 (shown in FIG.4) to perform a scan along the z-axis of a slide 420 at high magnification to identify the plane that contains the biomedical specimen of interest (e.g., a biopsy). For example, scanner 410 may correspond to one of scanners 101-109. Scanner 410 may acquire a series of images 430 along the z-axis using the same high magnification objective lens (e.g., 10x, 20x, or 40x) that will ultimately be used to acquire images used for the digitized slide. One or more attributes of the series of images 430 (e.g., color intensity) may be analyzed to estimate the plane containing the biopsy (or other biomedical specimen), as shown illustratively at process 440. For example, the biopsy plane may correspond to the z-axis coordinate that produces an image with the maximum color intensity. Still referring to FIGS.2 and 4, in some embodiments, biopsy plane estimation module 220 may perform biopsy plane estimation at x and y coordinates corresponding to reference point 338 (shown in FIG.3A-3B), e.g., a point determined by localization module 210 to be optimal for biopsy plane estimation. Performing biopsy plane estimation at reference point 338 may improve efficiency by ensuring that the biopsy plane estimation is performed at a location likely to contain a biomedical specimen. For example, the likelihood that biopsy plane estimation is repeated at different locations on the sample, or the likelihood that manual intervention is used to select or adjust the location for biopsy plane estimation, is reduced. Still referring to FIGS.2 and 4, in some embodiments, biopsy plane estimation module 220 may include one or more inline quality controls. For example, biopsy plane estimation module 220 may check that the biopsy plane estimation is performed at a location that contains a sufficient amount of the biomedical specimen to accurately perform biopsy plane estimation. When the check returns a negative result, one or more remedial actions may be taken, such as selecting a different location to perform biopsy plane estimation, generating an alert that insufficient biomedical specimen was detected, or the like. In some embodiments, biopsy plane estimation module 220 may output information associated with the results of the quality checks, 34 Attorney Docket No.1519-029PCT1 such as a numerical score, that can be used by subsequent modules in data flow 200. The information may be aggregated with information supplied by preceding modules in data flow 200 (e.g., localization module 210) to provide a running indicator of the quality of the slide digitization process, which may trigger further remedial actions (e.g., rescans or alerts) when the running indictor exceeds a predetermined threshold. Referring now to FIGS.2 and 5, first quality metric may include focus sampling quality metric; and determining second scanning parameter may include identifying a focal setting at which a selected point is in focus. As used herein, a “focus sampling quality metric” is a quality metric which assesses the performance of a focus sampling module. In some embodiments, focus sampling quality metric may include one or more composite measures used to evaluate the precision and reliability of focus sampling module’s ability to identify a focal setting at which a selected point is in focus. In other embodiments, focus sampling quality metric may include one or more quantifiable measures utilized by focus sampling module within the system designed to identify a focal setting at which a selected point is in focus. As depicted in FIGS.2 and 5, data flow 200 may include a focus sampling module 230. As used herein, a “focus sampling module” is an automated system for identifying a focal setting at which a selected point is in focus. Focus sampling module 230 may carry out focus optimization at two or more sample points 512 and 514 (shown in FIG.5) within the scanning grid (e.g., grid 334). Sample points 512 and 514 may be selected in relation to a reference point 520 (e.g., reference point 338). For example, sample points 512 and 514 may be positioned along a same row of the scanning grid as reference point 520. At each of sample points 512 and 514, focus sampling module 230 may instruct the scanner to determine optimal focus settings for subsequent imaging. For example, the optimal focus settings at sample points 512 and 514 may be determined using a variety of known techniques for auto-focusing, e.g. sharpness of the image. When the optimal focus settings at sample points 512 and 514 are different, focus sampling module 230 may determine a gradient (or other suitable functional relationship) of the focus settings such that optimal focus settings for the remaining points in the scanning grid may be estimated, e.g., by interpolation. Focus sampling may therefore provide an efficient technique for estimating the optimal focus settings throughout the scanning grid by performing optimization at a subset of the points in the grid and using gradients to estimate the optimal settings at the remaining points. 35 Attorney Docket No.1519-029PCT1 Still referring to FIGS.2 and 5, in some embodiments, focus sampling module 230 may include one or more inline quality controls. For example, focus sampling module 230 may check that the focus sampling is performed at locations that contain a sufficient amount of the biomedical specimen to accurately determine the focus settings. This may be performed by estimating the quality of specimen (e.g. thickness of specimen in z-direction), checking attributes of the specimen such as whether the specimen present on the glass slide corresponds to tissue versus annotation (e.g., pen marks), and confirming planarity of the slide estimated. Such locations may include, in non-limiting examples, fields of view of high resolution images of sections of a slide. When such a check returns a negative result, one or more remedial actions may be taken, such as selecting different locations to perform focus sampling, generating an alert that insufficient biomedical specimen was detected, or the like. In some embodiments, focus sampling module 230 may output information associated with the results of the quality checks, such as a numerical score, that can be used by subsequent modules in data flow 200. The information may be aggregated with information supplied by preceding modules in data flow 200 (e.g., modules 210-220) to provide a running indicator of the quality of the slide digitization process, which may trigger further remedial actions (e.g., rescans or alerts) when the running indictor exceeds a predetermined threshold. In some embodiments, focus sampling inline quality controls may include identifying locations where focus sampling errors are present. For example, locations which are out of focus may be identified, and remedial actions (such as rescans) may be taken for those areas. In some embodiments, such locations may include fields of view of high resolution images of sections of a slide. Referring now to FIGS.2 and 6A-6B, first quality metric may include z-stack acquisition quality metric; and determining second scanning parameter may include using the optical system, capturing a z-stack at a selected point. As used herein, a “z-stack acquisition quality metric” is a quality metric which assesses the performance of a z-stack acquisition module. As depicted in FIGS.2 and 6A-6B, data flow 200 may include a z-stack acquisition module 240. Based on information provided by preceding modules 210-240 (e.g., the scanning pattern, biopsy plane, and focus settings), z-stack acquisition module 240 instructs a scanner 610 (shown in FIGS.6A-6B) to acquire a set of images 620 (referred to as a z-stack) along the z-axis of a slide at high magnification. For example, scanner 610 may correspond to one of scanners 101-109. As shown in FIG.6B, the images in the z-stack may be acquired by sweeping the depth of field 36 Attorney Docket No.1519-029PCT1 using an objective lens of scanner 610. In this manner, features of the biomedical specimen (e.g., the tissue sample) that vary along the z-axis, such as bumps, ridges, and other 3-dimensional features, may come into focus in different layers of the z-stack. In some embodiments, z-stack acquisition module 240 may output an aggregate image based on the set of images in the z-stack. For example, as illustrated in FIG.6B at process 630, the aggregate image may correspond to a selected image from the z-stack that contains the most in-focus content among the images in the set. Other suitable techniques for aggregating the z-stack images may be used. Z-stack acquisition module 240 may acquire z-stack images at each x-y location in the sample to be scanned (e.g., as determined by localization module 210). Still referring to FIGS.2 and 6A-6B, in some embodiments, z-stack acquisition module 240 may include one or more inline quality controls. For example, z-stack acquisition module 240 may check that the amount of in focus image content in the z-stack exceeds a predetermined threshold, that the amount of debris or imaging artifacts in the images is below a predetermined threshold, that the levels of tissue folding and/or slide preparation artifacts are below a predetermined threshold, that the tissue is inside/outside the coverslip or the like, that the detected specimen corresponds to tissue rather than an annotation, or the like. When one or more of these checks return a negative result, one or more remedial actions may be taken, such as reacquiring the z-stack, adjusting the position of the z-stack (e.g., capturing images at a different position along the z-axis), adjusting the z stack size, generating an alert, or the like. Still referring to FIGS.2 and 6A-6B, in some embodiments, z-stack acquisition module 240 may track quality indicators associated with each z-stack as scanning proceeds throughout the scanning grid. For example, z-stack acquisition module 240 may count the locations in the scanning grid for which the quality checks produced negative results and may perform a remedial action when the count exceeds a predetermined threshold. In some embodiments, z- stack acquisition module 240 may generate and display a map indicating the quality of each z- stack in the scanning grid. Still referring to FIGS.2 and 6A-6B, in some embodiments, z-stack acquisition module 240 may output information associated with the results of the quality checks, such as a numerical score or a map indicating the quality of the z-stack scan at each location in the slide. This information can be used by subsequent modules in data flow 200. The information may be aggregated with information supplied by preceding modules in data flow 200 (e.g., modules 210- 37 Attorney Docket No.1519-029PCT1 230) to provide a running indicator of the quality of the slide digitization process, which may trigger further remedial actions (e.g., rescans or alerts) when the running indictor exceeds a predetermined threshold. Referring now to FIGS.2 and 7A-7B, performing a scan of a slide using stitching module 700 may include, using an optical system, capturing at least a second image as a function of second scanning parameter. In some embodiments, capturing at least a second image may include capturing a plurality of images, and stitching such images together. In some embodiments, performing a scan of a slide may include determining a stitching quality metric as a function of the at least a second image. In some embodiments, performing a scan of a slide may further include determining a third scanning parameter as a function of the stitching quality metric; using the optical system, capturing a second plurality of images as a function of the third scanning parameter; and combining the second plurality of images to create at least a third image. For example, if stitching quality metric is determined such that severe stitching related errors are identified, then additional images may be captured of an affected region, and such images may be used to assemble a corrected image. In some embodiments, image data from both at least a second image and subsequently captured image data is used to assemble at least a third image. As used herein, a “stitching quality metric” is a quality metric which assesses the performance of a stitching module. In some embodiments, stitching quality metric may include one or more composite measures used to evaluate the precision and reliability of stitching module’s ability to assemble a larger image from a plurality of images. In other embodiments, localization quality metric may include one or more quantifiable measures utilized by localization module within the system designed to assemble a larger image from a plurality of images. As depicted in FIGS.2 and 7A-7B, data flow 200 may include a stitching module 250. As used herein, a “stitching module” is an automated system for assembling a larger image from a plurality of images. Stitching module 230 aligns and stitches together (also referred to as pasting or fusing) adjacent images within the scanning grid 710 (shown in FIG.7A) to form an aggregated output image. In an ideal scenario, each pair of adjacent images acquired by the scanner (e.g., images 712 and 714) is aligned without observable gaps, overlaps, or offsets between the images, including in the x, y, and z directions. In practice, however, such defects may be present and may result in artifacts such as ghosting (e.g., regions that are blurred due to stitching errors). Another type of artifact that may arise due to stitching is the visual effect of 38 Attorney Docket No.1519-029PCT1 banding, as illustrated in FIG.7B. In an image with banding 722, artifacts such as vertical lines may appear along the axes corresponding to the scanning grid. Still referring to FIGS.2 and 7A-7B, in some embodiments, stitching module 250 may include one or more inline quality controls to address stitching errors, such as ghosting artifacts and displacement. For example, stitching module 250 may check that the presence of such artifacts in the output image is below a predetermined threshold. When the threshold is exceeded, one or more remedial actions may be taken, such as rescanning portions of the slide where the artifacts were present, or the like. Additional remedial actions may include triggering a non- inline stitching process. For example, a scanning process may be completed, and a subsequent stitching process may be started. In some embodiments, the inline quality checks may output information associated with the results of the quality checks, such as a numerical score indicating the level of stitching artifacts, that can be used by subsequent modules in data flow 200. The information may be aggregated with information supplied by preceding modules in data flow 200 (e.g., modules 210-240) to provide a running indicator of the quality of the slide digitization process, which may trigger further remedial actions (e.g., rescans or alerts) when the running indictor exceeds a predetermined threshold. Image without banding 724 depicts a non-limiting example of a correction which may be made. Returning to FIG.2, performing a scan may further include determining a second quality metric as a function of at least a first image; determining a combination quality metric as a function of first quality metric and second quality metric; determining a third scanning parameter as a function of the combination quality metric; and using optical system, capturing at least a third image as a function of second scanning parameter. Data flow 200 may include a whole slide assessment module 260. As used herein, a “whole slide assessment module” is an automated system for analyzing an output image generated by a stitching module. Whole slide assessment module 260 may perform a variety of quality checks on the output image and/or may synthesize the results of quality checks performed by modules 210-250. For example, whole slide assessment module 260 may perform a final quality check on the output image to trap out of focus, stitching, banding, missed specimen, z-stack shift errors (x,y shift across the image along the z plane of stack), or the like. When the check produces a negative result, one or more remedial actions may be taken such as rescanning portions of the slide where the errors were detected. In some embodiments, the slide may remain in the scanner during these quality checks, 39 Attorney Docket No.1519-029PCT1 such that taking a remedial action (e.g., rescanning a portion of the slide) involves little or no additional setup time. Still referring to FIG.2, in some embodiments, whole slide assessment module 260 may generate or compile a representation of the slide (e.g., a map or other suitable representation) that delineates portions of the output image in which artifacts persist but are within an acceptable threshold (e.g., portions in which errors are detected but not fully resolved by the quality controls identified above). The representation may enable the downstream application to handle the remaining artifacts in a suitable manner, e.g., by avoiding or applying inpainting techniques to the portions of the image that contain artifacts. Still referring to FIG.2, in some embodiments, performing a scan of a slide may further include algorithmically removing a banding error from an image such as at least a second image. In some embodiments, banding errors may be a product of non-uniform lighting across images and/or regions of images. Still referring to FIG.2, in some embodiments, an error may be flagged. For example, an error may be flagged based on a quality metric. In some embodiments, an error may be flagged as a function of a focus sampling quality metric. In some embodiments, an error may be flagged as a function of a stitching quality metric. In some embodiments, an error may be flagged if re- scanning a slide and/or a region of a slide does not sufficiently lead to a reduction in a quality metric. In some embodiment, prevalence and/or severity of errors may be tracked using a map. For example, a map may depict sections of an image and quality metrics associated with such sections of the image. For example, a map may include an overlay over an image which is color coded according to a particular quality metric and/or an aggregate measure of multiple quality metrics. In some embodiments, a map may be used to expedite manual quality control of an image. For example, a map may be output to a user using a display as described below in order to aid the user in a quality control process. Still referring to FIG.2, in some embodiments, system 100 may transmit one or more images and/or maps to an external device. Such an external device may include, in non-limiting examples, a phone, tablet, or computer. In some embodiments, such a transmission may configure the external device to display an image. Still referring to FIG.2, in some embodiments, system 100 may determine a visual element data structure. In some embodiments, system 100 may display to a user a visual element 40 Attorney Docket No.1519-029PCT1 as a function of visual element data structure. As used herein, a “visual element data structure” is a data structure describing a visual element. As non-limiting examples, visual elements may include at least a first image, at least a second image, at least a third image, a map, and elements of a GUI. Still referring to FIG.2, in some embodiments, a visual element data structure may include a visual element. As used herein, a “visual element” is a datum that is displayed visually to a user. In some embodiments, a visual element data structure may include a rule for displaying visual element. In some embodiments, a visual element data structure may be determined as a function of first image, second image, and/or hybrid image. In some embodiments, a visual element data structure may be determined as a function of an item from the list consisting of at least a first image, at least a second image, at least a third image, elements of a GUI, first scanning parameter second scanning parameter, third scanning parameter, first quality metric, and stitching quality metric. In a non-limiting example, a visual element data structure may be generated such that visual element depicting at least a second image is displayed to a user. Still referring to FIG.2, in some embodiments, visual element may include one or more elements of text, images, shapes, charts, particle effects, interactable features, and the like. As a non-limiting example, a visual element may include a touch screen button for setting magnification level. Still referring to FIG.2, a visual element data structure may include rules governing if or when visual element is displayed. In a non-limiting example, a visual element data structure may include a rule causing a visual element including a scanning parameter to be displayed when a user selects an image captured using that scanning parameter using a GUI. Still referring to FIG.2, a visual element data structure may include rules for presenting more than one visual element, or more than one visual element at a time. In an embodiment, about 1, 2, 3, 4, 5, 10, 20, or 50 visual elements are displayed simultaneously. For example, a plurality of images and/or GUI elements may be displayed simultaneously. Still referring to FIG.2, in some embodiments, apparatus may transmit visual element to a display such as output interface. A display may communicate visual element to user. A display may include, for example, a smartphone screen, a computer screen, or a tablet screen. A display may be configured to provide a visual interface. A visual interface may include one or more virtual interactive elements such as, without limitation, buttons, menus, and the like. A display 41 Attorney Docket No.1519-029PCT1 may include one or more physical interactive elements, such as buttons, a computer mouse, or a touchscreen, that allow user to input data into the display. Interactive elements may be configured to enable interaction between a user and a computing device. In some embodiments, a visual element data structure is determined as a function of data input by user into a display. Still referring to FIG.2, a variable and/or datum described herein may be represented as a data structure. In some embodiments, a data structure may include one or more functions and/or variables, as a class might in object-oriented programming. In some embodiments, a data structure may include data in the form of a Boolean, integer, float, string, date, and the like. In a non-limiting example, a quality metric data structure may include an int value representing a degree to which an image is in focus. In some embodiments, data in a data structure may be organized in a linked list, tree, array, matrix, tenser, and the like. In some embodiments, a data structure may include or be associated with one or more elements of metadata. A data structure may include one or more self-referencing data elements, which processor may use in interpreting the data structure. In a non-limiting example, a data structure may include “<date>” and “</date>,” tags, indicating that the content between the tags is a date. Referring now to FIG.8, a simplified diagram of a method 800 for inline quality control of slide digitization according to some embodiments is provided. According to some embodiments consistent with other figures, method 800 may be performed by processor 130 during the execution of inline quality control program identification program 150. For example, method 800 may be performed using one or more of modules 210-260. Still referring to FIG.8, at process step 810, an inline quality metric associated with digitizing a slide is evaluated. In some embodiments, the inline quality metric may correspond to one or more of the inline quality controls described previously with respect to FIGS.2-7B. For example, the inline quality metric may include a value that indicates the presence of defects in one or more slide images, such as debris, out-of-focus errors, stitching artifacts, misalignments, lack of a biomedical specimen, or the like. In some embodiments, the inline quality metric may correspond to an aggregate metric that is accumulated over multiple stages of the slide digitization process (e.g., multiple modules of data flow 200). The inline quality metric can be quantitative (e.g., a numerical score or a count), qualitative (e.g., good, fair, poor), or a combination thereof. Evaluating the inline quality metric may include applying a machine 42 Attorney Docket No.1519-029PCT1 learning model trained to identify defects and/or other image features relevant to inline quality control. Still referring to FIG.8, at process step 820, a remedial action is taken in response to a determination that the inline quality metric is within a predetermined range. For example, the inline quality metric may be within the predetermined range when its value exceeds a predetermined threshold, drops below a predetermined threshold, takes on a value identified as being associated with a defect (e.g., “poor” or “out-of-focus”), or the like. In some embodiments, the remedial action may include taking one or more of the remedial actions described previously with respect to FIGS.2- 7B. For example, the remedial action may include rescanning all or a portion of the slide, alerting a clinician, fixing an identified defect (e.g., applying image processing methods to reduce artifacts, inpainting one or more portions of the image using a generative machine learning model, or the like), generating a representation of the slide that indicates the presence or locations of defects for downstream applications, or the like. In some embodiments, the remedial action may be taken prior to removing the slide from the scanner, such that the set-up and/or take-down time associated with the remedial action is reduced, and manual intervention is likewise reduced. Still referring to FIG.8, in some embodiments, there may be multiple predetermined ranges associated with different remedial actions. For example, when the inline quality metric falls within a first predetermined range (e.g., a range indicative of the slide having an acceptable level of defects), the remedial action may include fixing the identified defects and/or identifying the presence or location of the defects for downstream processing stages. When the inline quality metric falls within a second predetermined range (e.g., a range indicative of the slide having an unacceptable level of defects), the remedial action may include initiating a rescan and/or alerting a clinician. Still referring to FIG.9, a box diagram depicting an exemplary embodiment of a system 900 for digitizing a slide is provided. System 900 may include processor 904, and memory 908 containing instructions 912 configuring processor 904 to perform one or more processes described herein. Computing device 916 may include processor 904, and memory 908, and may interact with other elements of system 900, such as by configuring elements of an optical system to capture images, configuring an actuator mechanism to move, configuring a user interface to display information and the like. System 900 may further include slide 920, such as a glass slide 43 Attorney Docket No.1519-029PCT1 containing a biological sample. Optical sensor 924 may be used to capture one or more images of slide 920. Actuator mechanism 928 may be used to move optical sensor 924, slide port 944, and/or slide 920 into correct positions for images to be captured according to desired parameters. System 900 may further include input interface 932, such as a mouse and keyboard, output interface 936 such as a screen and speaker system, and user interface 940 which includes input interface 932 and output interface 936. Computing device 916 may use settings according to first scanning parameter 948 to capture at least a first image 952. At least a first image 952 may be used to determine first quality metric 956. First quality metric 956 may include localization quality metric 960, biopsy plane estimation quality metric 964, focus sampling quality metric 968, and/or z-stack acquisition quality metric 972. First quality metric 956 may be used to determine second scanning parameter 976. Settings of second scanning parameter 976 may be used to capture at least a second image 980. Stitching quality metric 984 may be determined as a function of at least a second image 980, such as in a case in which second image 980 is assembled from a plurality of images each covering narrower regions of slide 920. Third scanning parameter 988 may be determined as a function of stitching quality metric 984, and at least a third image 992 may be captured using settings of third scanning parameter 988. One or more of at least a first image 952, at least a second image 980, and at least a third image 992 may be displayed using user interface 940. In a non-limiting example, system 900 may capture a macro image which includes in frame the entirety of a biological sample which a user desires a high resolution image of. In this example, system 900 may use this macro image to identify one or more parameters for capturing subsequent images as described herein; such parameters may include a region within the macro image which contains the sample, a row within this region in which the sample is present, a focal distance at one or more points are in focus, and the like. In this example, system 900 may use such parameters to capture a plurality of images and may assemble a high resolution image from this plurality of images. This plurality of images may be checked for errors, such as stitching errors, and these errors may be corrected. A final image may be displayed to a user using a user interface. Referring now to FIG.10, an exemplary embodiment of a method 1000 of digitizing a slide is illustrated. One or more steps if method 1000 may be implemented, without limitation, as described with reference to other figures. One or more steps of method 1000 may be implemented, without limitation, using at least a processor. Method 1000 may include 44 Attorney Docket No.1519-029PCT1 performing a scan of a slide by capturing at least a first image and capturing at least a second image. Still referring to FIG.10, in some embodiments, performing a scan of a slide may include identifying a first scanning parameter 1005. Still referring to FIG.10, in some embodiments, performing a scan of a slide may include using an optical system, capturing the at least a first image as a function of the first scanning parameter 1010. Still referring to FIG.10, in some embodiments, performing a scan of a slide may include determining a first quality metric as a function of the at least a first image 1015. In some embodiments, the first quality metric includes a localization quality metric; and determining the second scanning parameter includes identifying at least a region of interest of the slide. In some embodiments, the first quality metric includes a biopsy plane estimation quality metric; and determining the second scanning parameter includes identifying a plane within a region of interest which contains a biological specimen. In some embodiments, the first quality metric includes a focus sampling quality metric; and determining the second scanning parameter includes identifying a focal setting at which a selected point is in focus. In some embodiments, the first quality metric includes a z-stack acquisition quality metric; and determining the second scanning parameter includes, using the optical system, capturing a z-stack at a selected point. Still referring to FIG.10, in some embodiments, performing a scan of a slide may include determining a second scanning parameter as a function of the first quality metric 1020. Still referring to FIG.10, in some embodiments, performing a scan of a slide may include using the optical system, capturing the at least a second image as a function of the second scanning parameter 1025. In some embodiments, the at least a first image includes a macro image; and capturing the at least a second image includes, using the optical system, capturing a first plurality of images; and combining the first plurality of images to create the at least a second image. Still referring to FIG.10, in some embodiments, performing the scan further includes determining a stitching quality metric as a function of the at least a second image; determining a third scanning parameter as a function of the stitching quality metric; using the optical system, capturing a second plurality of images as a function of the third scanning parameter; and combining the second plurality of images to create at least a third image. In some embodiments, 45 Attorney Docket No.1519-029PCT1 performing the scan further includes capturing the at least a first image at a first location and capturing the at least a second image at a second location. In some embodiments, performing the scan further comprises determining a second quality metric as a function of the at least a first image; determining a combination quality metric as a function of the first quality metric and the second quality metric; determining a third scanning parameter as a function of the combination quality metric; and using the optical system, capturing at least a third image as a function of the second scanning parameter. In some embodiments, performing the scan further includes algorithmically removing a banding error from the at least a second image. In some embodiments, upon detection of a banding error and/or banding errors of sufficient severity, a remedial action may be taken. Remedial actions may include, in non-limiting examples, pausing a scanner, triggering a cleaning mechanism in order to remove and/or reduce banding errors, and/or switching to alternative optical components. In some embodiments, one or more aspects of a system described herein may be implemented as described in U.S. Pat. App. No.18/384,840, filed on October 28, 2023, and titled “APPARATUS AND METHODS FOR SLIDE IMAGING”, the entirety of which is hereby incorporated by reference. For example, localization module, biopsy plane estimation module, focus sampling module, z-stack acquisition module, stitching module, and/or whole slide assessment module may be implemented as described in the incorporated patent application. At a high level, aspects of the present disclosure are directed to systems and methods for digitization of tissue slides based on associations among serial sections. In an embodiment, the digitization of tissue slides based on associations among serial sections may occur via a system including at least a computing device, having a processor and a memory, and a scanner configured to scan a slide and send a digitized image of the slide to the computing device. In another aspect a method for digitizing tissue slides based on associations among serial sections may include receiving a candidate tissue map associated with a candidate tissue section, receiving a reference tissue map associated with a reference tissue section, aligning the candidate tissue map tot the reference tissue map, comparing the aligned candidate tissue map to the reference tissue map, and generating a regenerated candidate tissue map as a function of the reference tissue map. Aspects of the present disclosure can be used to enable the acquisition of tissue content while avoiding debris, bubbles, annotations, and background stain. Aspects of the present 46 Attorney Docket No.1519-029PCT1 disclosure can also be used to improve the accuracy of scanning of tissue specimen having various types, sizes, fragments, and/or stain level. This is so, at least in part, because when serial section slides are processed as a group, the analysis benefits from the minimization of issues related to individual slide tissue assessment. Exemplary embodiments illustrating aspects of the present disclosure are described below in the context of several specific examples. Referring now to FIG.11, an exemplary embodiment of a system for digitization of tissue slides based on associations among serial sections is illustrated. A system for digitization of tissue slides based on associations among serial sections may include at least a computing device 1104 and scanner 1124 configured to scan a slide and send a digitized image of the slide to computing device 1104. Wherein computing device 1104 further includes processor 1108 and memory 1112 communicatively connected to processor 1108, wherein memory 1112 has stored instructions configuring processor 1108 to retrieve a candidate tissue map associated with a candidate tissue section, retrieve a reference tissue map associated with a reference tissue section, align candidate tissue map to reference tissue map, compare aligned candidate tissue map to reference tissue map, and generate a regenerated candidate tissue map as a function of reference tissue map. This may be implemented in part, without limitation, as disclosed in U.S. App. No.18/428,823, filed on 01/31/2024 and entitled “SYSTEMS AND METHODS FOR VISUALIZATION OF DIGITIZED SLIDES” the entirety of which is incorporated herein by reference. In some embodiments, computing system may further include one or more storage devices. For example, and without limitation, scanned slides data repository 1116 and/or regenerated slides data repository 1120. “Digitization,” as used throughout this disclosure, is the conversion of text, pictures, and/or sound into a digital form that can be processed by a computer. This may be implemented, without limitation, as disclosed in U.S. App. No. 63/466,950, filed on 5/16/2023 and entitled “SYSTEMS AND METHODS FOR INLINE QUALITY CONTROL OF SLIDE DIGITIZATION” the entirety of which is incorporated herein by reference. Further, this may be enhanced, without limitation, as disclosed in U.S. App. No.18/227,155, filed on 07/27/2023 and entitled “METHOD AND AN APPARATUS FOR INLINE IMAGE SCAN ENRICHMENT” the entirety of which is incorporated herein by reference. As used throughout this disclosure, “serial sections,” refer to any series of sections cut in sequence from a prepared specimen. Additionally, a “repository,” as used in this disclosure, is a receptable where things are or may be stored. In the instance of scanned slides data repository 47 Attorney Docket No.1519-029PCT1 1116 and/or regenerated slides data repository 1120, each repository respectfully, may store and or be accessed for the information it holds. With further reference to FIG.11, system 1100 may include device 1104 and scanner 1124 communicatively connected to device 1104. In an embodiment, device 1104 may include a computing device and/or system in any embodiment as described throughout this disclosure. As a nonlimiting example, device 1104 may include a personal computer, an on-premises server, a cloud-based server, and/or the like. In an embodiment, scanner 1124 may include devices and/or systems used to digitize slides containing biomedical specimens, for example tissue samples. Such devices and/or systems may be any device and/or system as described throughout this disclosure. For example, and without limitation scanner 1124 may include digital cameras, digital microscopes, digital pathology scanners, and/or the like. Device 1104 and scanner 1124 may communicate via a wired and/or wireless link, and/or via a network, including but not limited to one or more local area networks (LANs), wide area networks (WANs), wired networks, wireless networks, the Internet, and/or the like. Alternatively, in some embodiments, device 1104 may be directly connected to and/or integrated within scanner 1124. Each of these descriptions are exemplary and may be substituted with any other embodiment as described in further detail throughout this disclosure. Still referring to FIG.11, scanner 1124 may be configured to capture a digital image, which may be referred to throughout this disclosure as digitized slide 1128. Digitized slides 1128 may include any image produced using a digital camera and that may be stored as an electronic file. Some examples of this may include binary, grayscale, color, and/or multispectral images. Following digitization of the slide, scanner 1124 may send digital image as a digital file and/or data to device 1104 via communicative link between scanner 1124 and device 1104. With further reference to FIG.11, device 1104 may include processor 1108 coupled to memory 1112. Memory 1112 may store instructions and/or data configuring processor 1108 to perform specific operations. When operations are executed by processor 1108, processor 1108 may perform operations associated with the inline registration of serial sections and more specifically the generation of maps of tissue regions. Processor 1108 may also be in communication with scanned slide data repository 1116 and/or regenerated slides data repository 1120. Processor 1108 may be further configured to read, write, and/or manage data stored in scanned slides data repository 1116 and regenerated slides data repository 1120. Scanned slides 48 Attorney Docket No.1519-029PCT1 data repository 1116 stores data of digitized slides 1128 from scanners, such as scanner 1124. Stored data may include image data and/or meta data, such as case identification numbers and/or block identification numbers corresponding to the image data. In some embodiments, scanned slides data repository 1116 and/or regenerated slides data repository 1120 may be located in another device and/or a device that is separate from device 1104. In an embodiment, device 1104 may be configured to communicate with the separate device via a wired and/or wireless link, and/or via a network including one or more local area networks (LANs), wide area networks (WANs), wired networks, wireless networks, the Internet, and/or the like. Continuing to reference FIG.11, in an embodiment slide digitization may include the retrieval of candidate slides and reference slides identified based on stain information of slides under assessment. In some embodiments, identification of candidate slides and/or reference slides may automatically be accomplished real-time alongside the digitization process by an inline computer program executed by processor 1108. An “inline computing program,” refers to a computing term where code or data is inserted directly into its appropriate place within a larger block of code, rather than being called from a separate location. Alternatively, in some embodiments, identification of candidate slides and/or reference slides may be manually performed by a user. The identified candidate slide may be associated with information sufficient to identify one or more other slides from the same serial section set, which may be used as the reference slide. For example, and without limitation, candidate slide may have a corresponding case identification number and block identification number. Based on the case identification, as well as stain information of slides having the corresponding case identification number and the block identification number, another slide corresponding to the case identification number and the block identification number may be identified as a reference slide. Further referencing FIG.11, in some embodiments identified reference slides may be an H&E-stained slide. In contrast, identified candidate slide may be a non-H&E-stained slide. For example, and without limitation non-H&E-stained slides may be stained with Immunohistochemistry (IHC). IHC staining combines anatomical, immunological, and biochemical techniques to image discrete components in tissues by using appropriately labeled antibodies to bind specifically to their target antigens in situ. IHC staining makes it possible to visualize and document high-resolution distribution and localization of specific cellular components within cells and within their proper histological context. In some embodiments, 49 Attorney Docket No.1519-029PCT1 identified reference slide may be a non-H&E-stained slide. When multiple candidate slides are identified, there may be any combination of a certain number of non-H&E-stained slides and a certain number of H&E-stained slides. Continuing to reference FIG.11, processor 1108 may be configured to retrieve image data of reference slide and image data of candidate slide from scanned slides data repository 1116 to memory 1112. The described retrieval operation by processor 1108 may be based on the case identification number and the block identification number of candidate slide, which have been supplied to processor 1108. Using image data of reference slide and image data of candidate slide, processor 1108 generates reference tissue map file 1132 and candidate tissue map file 1136, respectively. Reference tissue map file 1132 may contain data of the tissue map of candidate slide. With further reference to FIG.11, in some embodiments, a user may identify a reference serial section and at least one non-reference serial action, otherwise described as a candidate serial section, on the same slide based on information such as shape and/or other attributes of the serial sections. The slide is also referred to as an intra-serial sections slide. In turn, reference tissue map file 1132 may be generated by processor 1108 from image data of reference serial section and candidate tissue map file 1136 may be generated by processor 1108 from image data of candidate serial section. Continuing to reference FIG.11, processor 1108 may additionally be configured to align candidate tissue map file 1136 with reference tissue map file 1132. During this alignment process, candidate tissue map file 1136 may be registered by processor 1108 with the reference map file by leveraging information such as similarities of shape and size between mapped serial sections. For example, and without limitation, processor 1108 may execute a registration module containing instructions stored in memory 1112 to process candidate tissue map and reference tissue map. Processor 1108 orients candidate tissue map, which may be stored in candidate tissue map file 1136, to maximize the overlap between regions of interest identified in candidate tissue map and reference tissue map, which may be stored in reference tissue map file 1132. Features used to align tissue maps may include tissue boundaries, contours, and/or other features that are visible in a slide, even in cases where such features may appear faint. The registration module may align one or more features that are common between candidate tissue map and reference tissue map. This may be accomplished by maximizing the overlap between common features in 50 Attorney Docket No.1519-029PCT1 two or three dimensions in both tissue maps. Common features may include features that are detectable, such as without limitation, features discernable using automated computer vision techniques. Maximizing overlap may be achieved by applying one or more matrix transformations, such as, without limitation, rotating, translating, scaling and/or skewing, to candidate tissue map and/or reference tissue map. With further reference to FIG.11, processor 1108 may be further configured to compare candidate tissue map file 1136 with reference tissue map file 1132 regarding properties such as shapes and sizes between mapped serial sections. In an embodiment, when a difference in compared properties is located as result of the comparison, differences are then utilized as correction factors to generate regenerated candidate tissue maps. In some embodiments, this process may include an initial step of generating a reference binary mask from reference tissue map and a candidate binary mask from candidate tissue map. These binary masks may identify the presence or absence of tissue across the slide in contrast to tissue maps which may provide a more detailed representation of the features on the slide. For example, and without limitation, binary masks may appear in a black and white format in contrast to RGB color format of features contained on a given slide. Reference binary mask and candidate binary mask may be aligned based on results of the alignment process as described above. For example, and without limitation, by applying matrix transformations determined by the registration module. Once aligned, adjustments may be made to candidate binary mask based on reference binary mask, resulting in a corrected candidate binary mask. As a nonlimiting example of this, a region of candidate binary mask originally identified as having no tissue present may be identified as having tissue in corrected candidate binary mask when the reference binary mask indicates that tissue is present in the region. Similarly, regions of candidate binary mask originally identified as having tissue present may be identified as having no tissue in corrected candidate binary mask when reference binary mask indicates that no tissue is present in the region. The corrected candidate binary mask may then be applied onto candidate tissue map, and candidate tissue map may be regenerated accordingly with corrections. Consequently, in some embodiments, the area of regenerated candidate tissue map where tissue is determined to be present is substantially equal to the corresponding area of reference tissue map. Continuing to reference FIG.11, these and other factors may be streamlined by instantiation of a machine learning module and/or a neural network. Training data that may be 51 Attorney Docket No.1519-029PCT1 used to train machine-learning model and/or neural network may include exemplary input data, such as without limitation, digitized slide 1128 data, such as without limitation class identification numbers, and/or block identification numbers, and/or the like, candidate tissue map data, reference tissue map data, regenerated candidate tissue map data, binary maps of any tissue map data, and/or the like, where each such example may be correlated to additional exemplary output data such as, without limitation, regenerated candidate tissue map data, and/or the like. Training of the model and/or network may take place either at device 1104 and/or remotely. In the latter case, the model and/or network may be deployed at or by device 1104 in any manner as described in this disclosure. In some embodiments, the machine-learning model and/or neural network may be trained remotely and then transmitted device 1104, the model and/or network may be deployed at or device 1104 in any manner described in this disclosure. Additionally, in some embodiments, the machine-learning model and or neural network may be updated to device 1104, the model and/or network may be deployed at or by device 1104 in any manner as described in this disclosure. The machine-learning model and/or network may be deployed/instantiated once trained in any form as described within this disclosure. Feedback from the deployment of the machine-learning model and/or neural network may be turned into new training data, which may be stored either locally and/or transmitted to another device and used for retraining of the model and/or network. Retraining may be administered either remotely or at device 1104. Following retraining of the model and/or network, redeployment/instantiation may be accomplished at or by device 1104 in any manner as described within this disclosure. Further referencing FIG.11, a regenerated candidate tissue map may be stored in a regenerated candidate tissue map file 1140, which may be used by downstream applications such as further processing, data storage, and/or high-resolution scanning with large magnification multiples. Regenerated slides data repository 1120 may store data of regenerated candidate tissue map. In an embodiment, as described throughout this disclosure, this may be referred to as regenerated candidate tissue map file 1140. In some embodiments, binary masks and tissue maps generated during the process of slide digitization are configured to be low resolution without large magnification multiples. With continued reference to FIG.11, generating regenerated candidate tissue maps in an inline manner, as described above, may have various advantages. For example, candidate tissue map may be generated based on an image of a slide captured by scanner 1124 at a low 52 Attorney Docket No.1519-029PCT1 magnification, such as 1x magnification, and regenerated candidate tissue map may be generated prior to removing slide from scanner 1124. In turn, scanner 1124 may be instructed to carry out higher resolution scanning, such as 40x magnification, of specified regions of interest determined based on regenerated candidate tissue map. In this manner, various inefficiencies may be avoided, such as undesirably carrying out high resolution scanning based on original defective candidate tissue map, resulting in potentially performing high resolution scanning on the wrong regions of slide and/or missing regions that it would have been desirable to scan at high resolution. Additionally, one may avoid untimely detection of errors, such as detection of an error after slide has been removed from scanner 1124 and placed back into storage, resulting in potentially significant delays and additional manual processing. Continuing to reference FIG.11, system 1100 includes computing device 1104. Computing device 1104 includes processor 1108 communicatively connected to memory 1112. As used in this disclosure, “communicatively connected” means connected by way of a connection, attachment or linkage between two or more relata which allows for reception and/or transmittance of information therebetween. For example, and without limitation, this connection may be wired or wireless, direct or indirect, and between two or more components, circuits, devices, systems, and the like, which allows for reception and/or transmittance of data and/or signal(s) therebetween. Data and/or signals therebetween may include, without limitation, electrical, electromagnetic, magnetic, video, audio, radio and microwave data and/or signals, combinations thereof, and the like, among others. A communicative connection may be achieved, for example and without limitation, through wired or wireless electronic, digital or analog, communication, either directly or by way of one or more intervening devices or components. Further, communicative connection may include electrically coupling or connecting at least an output of one device, component, or circuit to at least an input of another device, component, or circuit. For example, and without limitation, via a bus or other facility for intercommunication between elements of a computing device. Communicative connecting may also include indirect connections via, for example and without limitation, wireless connection, radio communication, low power wide area network, optical communication, magnetic, capacitive, or optical coupling, and the like. In some instances, the terminology “communicatively coupled” may be used in place of communicatively connected in this disclosure. 53 Attorney Docket No.1519-029PCT1 Further referring to FIG.11, Computing device 1104 may include any computing device as described in this disclosure, including without limitation a microcontroller, microprocessor, digital signal processor (DSP) and/or system on a chip (SoC) as described in this disclosure. Computing device 1104 may include, be included in, and/or communicate with a mobile device such as a mobile telephone or smartphone. Computing device 1104 may include a single computing device operating independently, or may include two or more computing device operating in concert, in parallel, sequentially or the like; two or more computing devices may be included together in a single computing device or in two or more computing devices. Computing device 1104 may interface or communicate with one or more additional devices as described below in further detail via a network interface device. Network interface device may be utilized for connecting computing device 1104 to one or more of a variety of networks, and one or more devices. Examples of a network interface device include, but are not limited to, a network interface card (e.g., a mobile network interface card, a LAN card), a modem, and any combination thereof. Examples of a network include, but are not limited to, a wide area network (e.g., the Internet, an enterprise network), a local area network (e.g., a network associated with an office, a building, a campus or other relatively small geographic space), a telephone network, a data network associated with a telephone/voice provider (e.g., a mobile communications provider data and/or voice network), a direct connection between two computing devices, and any combinations thereof. A network may employ a wired and/or a wireless mode of communication. In general, any network topology may be used. Information (e.g., data, software etc.) may be communicated to and/or from a computer and/or a computing device. Computing device 1104 may include but is not limited to, for example, a computing device or cluster of computing devices in a first location and a second computing device or cluster of computing devices in a second location. Computing device 1104 may include one or more computing devices dedicated to data storage, security, distribution of traffic for load balancing, and the like. Computing device 1104 may distribute one or more computing tasks as described below across a plurality of computing devices of computing device, which may operate in parallel, in series, redundantly, or in any other manner used for distribution of tasks or memory between computing devices. Computing device 1104 may be implemented, as a non-limiting example, using a “shared nothing” architecture. 54 Attorney Docket No.1519-029PCT1 With continued reference to FIG.11, computing device 1104 may be designed and/or configured to perform any method, method step, or sequence of method steps in any embodiment described in this disclosure, in any order and with any degree of repetition. For instance, computing device 1104 may be configured to perform a single step or sequence repeatedly until a desired or commanded outcome is achieved; repetition of a step or a sequence of steps may be performed iteratively and/or recursively using outputs of previous repetitions as inputs to subsequent repetitions, aggregating inputs and/or outputs of repetitions to produce an aggregate result, reduction or decrement of one or more variables such as global variables, and/or division of a larger processing task into a set of iteratively addressed smaller processing tasks. Computing device 1104 may perform any step or sequence of steps as described in this disclosure in parallel, such as simultaneously and/or substantially simultaneously performing a step two or more times using two or more parallel threads, processor cores, or the like; division of tasks between parallel threads and/or processes may be performed according to any protocol suitable for division of tasks between iterations. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which steps, sequences of steps, processing tasks, and/or data may be subdivided, shared, or otherwise dealt with using iteration, recursion, and/or parallel processing. Now referring to FIG.12, exemplary serial sections in various cases are shown. During the process of slide digitization, multiple serial sections may be assessed for a patient case and/or tissue block of a patient case. In this embodiment, these serial sections are therefore not independent from each other. Rather, the associations among these serial sections may give rise to patterns, trends, and/or other information that may be leveraged in the slide digitization process. For instance, such information can be utilized to correct a serial section under assessment which may have been contaminated. With further reference to FIG.12, first serial section set 1204 and a second serial section set 1208 may be related to a first patient case and thereby assigned a unique case identification number associated with the first patient case. The first serial section set 1204 may be acquired from a first tissue block of the first patient case and the second serial section set 1208 may be acquired from a second tissue block of the first patient case. Additionally, in some embodiments, a third serial section set 1212 may be related to a second patient case and thereby assigned another unique case identification number associated with the second patient case. The third 55 Attorney Docket No.1519-029PCT1 serial section set 1212 may be acquired from a first tissue block of the second patient case. The first serial set 1204 may include a first stain slide 1216, a second stain slide 1220, and a third stain slide 1224. The second serial section set 1208 may include a fourth stain slide 1228 and a fifth stain slide 1232. The third serial section set 1212 may include a sixth stain slide 1236 and a seventh stain slide 1240. Continuing to reference FIG.12, in determining the first tissue block of the first patient case, the first stain slide 1216 may be identified as a reference slide among the first serial section set 1204 that may be used for the slide digitization process described in the present disclosure. In some examples, the first stain slide 1216 identified as a reference slide may be an H&E-stained slide. When a group of slides are processed together, the reference slide may provide information for assessing other slides with other stains. For example, and without limitation, the second stain slide 1220 and the third stain slide 1224 may be utilized as candidate slides. In further reference to FIG.12, in some embodiments, a non-H&E-stained slide may be identified and used as a reference slide for processing a group of slides. For example, and without limitation, when determining the second tissue block of the first patient case, the second serial section set 1208 is processed as a group for the slide digitization process described in the present disclosure. In the second serial section set 1208, the fourth stain slide 1228 may be a non-H&E-stained slide, which may be identified and used as a reference slide for assessing the fifth stain slide 1232. The fifth stain slide 1232 may be a candidate slide in the second serial section set 1208. Still referencing FIG.12, in some embodiments, multiple serial sections may be placed on the same slide, which is referred to as an intra-serial section slide. For example, and without limitation, the sixth stain slide 1236 in the third serial section set 1212 is an intra-serial section slide. Alternatively, in embodiments that only one serial section is placed on a slide, the embodiment is referred to as an inter-serial section slide. Now referring to FIG.13, an exemplary method of digitization of tissue slides based on associations among serial sections may include receiving a candidate tissue map associated with a candidate tissue section 1305, receiving a reference tissue map associated with a reference tissue section 1310, aligning the candidate to the reference tissue map 1315, comparing the aligned candidate tissue map to the reference tissue map 1320, and generating a regenerated candidate tissue map as a function of the reference tissue map 1325. The receipt of the candidate 56 Attorney Docket No.1519-029PCT1 tissue map and the reference tissue map 1310 may flow from other methods and/or processes as described here within and/or from an independent process and/or method. These methods and/or processes may include other programs and or artificial intelligence programs configured to replicate the manual processing of slide analysis and/or digitization processes. Continuing to reference FIG.13, The alignment of the candidate tissue map to reference tissue map 1315 may further generate a reference-aligned candidate tissue map. In some embodiments, the process of such alignment may apply image registration techniques. By aligning one or more features such as tissue boundaries and/or contours shared by the candidate tissue map and the reference tissue map, the registration is performed. Alignment may occur in two and/or three dimensions in order to maximize the overlap of common features shared by both maps. Common features may include features that are detectable in both tissue maps. For example, by way of automated computer vision techniques as described in further detail throughout this disclosure. Further referencing FIG.13, comparison of the aligned candidate tissue map to the reference tissue map 1320 may further include generating a reference binary mask from the reference tissue map and a candidate binary mask from the candidate tissue map, aligning the reference binary mask and the candidate binary mase based on the results of the alignment process, adjusting the candidate binary mask based on the reference binary mask, resulting in a corrected binary mask, and applying the corrected candidate binary mask onto the candidate tissue map. This process facilitates comparison between substantially the same tissue-containing regions of the candidate tissue map and reference tissue map. With further reference to FIG.13, method 1300 may include generating a regenerated candidate tissue map 1325 based on the differences between the candidate tissue map and the reference tissue map identified within the comparison phase of the method. Otherwise stated as generating a regenerated candidate issue map as a function of the reference tissue map 1325. The identified differences within the comparison phase may be used as correction factors to adjust the candidate tissue map. Continuing to refer to FIG.13, a method for digitization of tissue slides based on associations among serial sections may further include identifying a candidate serial section from at least one stain type, identifying a reference serial section, generating the reference tissue map in response to scanning the reference serial section, and generating the candidate tissue map in 57 Attorney Docket No.1519-029PCT1 response to scanning the candidate serial section. In an embodiment, these additional methods and/or processes may be completed prior to receiving a candidate tissue map associated with the candidate tissue section. A candidate serial section may be identified based on stain information such as stain types. The candidate serial section is associated with identifying information such as a related pair of a case identification number and a block identification number. Likewise, a reference serial section may be identified based on identifying information such as the case identification number and the block identification number. In some embodiments, the candidate serial section and the reference serial section are on different slides. In other cases, the candidate serial section and the reference serial section are on the same slide. By identifying multiple sub- components on the same slide of repeating patterns, a slide with multiple serial sections may be identified. In some embodiments, the slides used for serial sections can be H&E slides, non-H&E slides, or any combination thereof. Generation of a reference tissue map in response to scanning the reference serial section may be accomplished by scanning the reference serial section with scanner and outputting a digitized slide. Likewise, generation of the candidate tissue map may occur through scanning of the candidate serial section resulting in another digitized slide. In some embodiments, based on aggregation of sub-components of a generated tissue map on a slide, the presence of a repeating pattern may indicate that multiple serial sections are present on the same slide. These processes, although stated in phased terms, may occur simultaneously and/or near simultaneously with other methods as described throughout this disclosure. Now referring to FIG.14, illustrated are serial sections on different slides at various stages during an exemplary process of slide digitization 1400. Prior to generation of candidate tissue map 1416 and reference tissue map 1412, a candidate binary mask 1408 and a reference binary mask 1404 are resulted from the scanning of serial sections on their respective slides. The candidate binary mask 1408 is initially wrongly assessed due to the presence of debris and/or lack of stain intensity of a portion of the tissue. Alternatively, the reference binary mask 1404 appears to be correctly assessed of tissue shape and size by an inline automatic computer program. When the inline registration is completed, the candidate tissue map 1416 is aligned with the reference tissue map 1420. The reference binary mask 1404 is used to correct the candidate binary mask 1408. As a result, an aligned candidate tissue map 1424 is generated. The aligned candidate tissue map 1424, which is based on the corrected candidate binary mask 1404, may then be used to regenerate the candidate tissue map. Following the process of regeneration 58 Attorney Docket No.1519-029PCT1 of the candidate tissue map, the regenerated candidate tissue map 1432 appears to represent tissue with similar shape and size as the reference tissue map 1428. Alternatively, the orientation and the relative location on the glass slide appears to be different between the regenerated candidate tissue map 1432 and the reference tissue map 1428. Now referring to FIG.15, illustrated is serial sections on the same slide at various stages during an example process of slide digitization 1500, in accordance with some embodiments. In the illustrated example, an intra-serial section slides 1504 may include a reference serial section 1508 and a candidate serial section 1512 on the same slide. An “intra-serial section slide,” as used throughout this disclosure is a slide that has more than one serial section placed on it. The rest of the slides as illustrated include serial slides at various points of the method as described above and throughout this disclosure. An intra-serial binary mask section slide is shown and may include a binary mask 1516 of a reference serial section 1520 and a candidate serial section 1524 on the same slide. The repeating serial section 1528 additionally may include a reference serial section 1532 and a candidate serial section 1536. Lastly, the regenerated candidate serial section map 1540 may include a reference serial section 1544 and a candidate serial section 1548. Continuing to reference FIG.15, in some embodiments, no slide is manually labeled as an intra-seral section slide before the process of slide digitization. Instead, an inline registration process of an intra-serial section slide is performed by a processor with automatic computer programs. In the first phase of the registration of intra-serial section slide, the inline registration program finds all repeating matches of serial sections matching features of the reference template, such as shapes and/or sizes. In the second phase of the registration of intra-serial section slide, the inline registration program expands the reference template to be applied onto all matching serial sections found in the first phase. When completing the first phase and the second phase, the slide being assessed is registered as an intra-serial section slide. With further reference to FIG.15, the information about the count and location of serial sections on an intra-serial section slide may be utilized in the process of digitizing intra-serial section slide. In some embodiments, a user may input parameters and/or instructions regarding the count and location of serial sections through a user interface. In some embodiments, such information may be detected by an inline algorithm configured to identify and/or locate sub- sections of tissue and/or sub-components of serial sections having similarities. Executing the inline algorithm stored in a memory device, a processor assesses the content of every slide to 59 Attorney Docket No.1519-029PCT1 determine whether there are repeated fragments present. In an embodiment where repeating fragments are present on a slide, they may be identified by the processor as an intra-serial sections slide. An intra-serial sections slide may also be referred to as an intra-serial fragment slide. The intra-serial section slide may be H&E-stained and/or non-H&E-stained. Continuing to reference FIG.15, during the process of digitizing an intra-serial section slide, one of the serial sections may be identified as a reference serial section. This determination may be determined by the processor by way of locating repeating patterns. Other serial sections on the same slide may then be checked against the reference serial section. In turn, the corresponding tissue maps are corrected in view of the reference serial section. In FIG.15 specifically, a reference serial section is assessed by the processor to identify and recover the missing tissue fragment from a candidate serial section. FIG.16 illustrates an example method 1600 for the inline registration of serial sections during slide digitization, according to some embodiments. In the illustrated example, the method 1600 includes steps 1602, 1604, 1606, 1608, 1610, 1612, and 1614. In some embodiments, the example method is performed with the system 1200 illustrated and described in FIG.12. At step 1602, a candidate serial section is identified based on stain information such as stain types. The candidate serial section is associated with identifying information such as a pair of a case identification number and a block identification number. At step 1608, the processor is configured to generate a candidate tissue map in response to scanning the candidate serial section. For instance, the candidate serial section is scanned by the scanner resulting another digitized slide. In some examples, based on aggregation of subcomponents of a generated tissue map on a slide, the presence of a repeating pattern indicates that multiple serial sections are identified on the same slide. At step 1610, the processor aligns the candidate tissue map to the reference tissue map, thereby generates a reference-aligned candidate tissue map. In some examples, the process of such alignment applies image registration techniques. By aligning one or more features such as tissue boundaries and/or contours that are common between the candidate tissue map and the reference tissue map, e.g., by maximizing the overlap between the common features in two or three dimensions, the registration is performed. In general, the common features are those that are detectable (e.g., discernable using automated computer vision techniques) in both tissue maps. Maximizing the overlap may be achieved by applying one or more matrix transformations, 60 Attorney Docket No.1519-029PCT1 such as rotating, translating, scaling and/or skewing, to the candidate tissue map and/or the reference tissue map. At step 1612, the processor compares the aligned candidate tissue map to the reference tissue map, thereby identifying differences between the aligned candidate tissue map and the reference tissue map. The identified differences may in turn be used to As discussed in further detail above, comparing the aligned candidate tissue map to the reference tissue map may include: (a) generating a reference binary mask from the reference tissue map and a candidate binary mask from the candidate tissue map, (b) aligning the reference binary mask and the candidate binary mask based on the results step 1610, (c) adjusting the candidate binary mask based on the reference binary mask, resulting in a corrected candidate binary mask, and (d) applying the corrected candidate binary mask onto the candidate tissue map. This process facilitates comparison between substantially the same tissue-containing regions of the candidate tissue map and reference tissue map. At step 1614, the processor generates a regenerated candidate tissue map at step 1614 based on the differences between the candidate tissue map and the reference tissue map identified at step 1612. For example, the identified differences may be used as correction factors to adjust the candidate tissue map. At a high level, aspects of the present disclosure are directed to inline image optimization during the scanning process. In an embodiment, apparatus and methods may include utilizing machine-learning to identify focal points for documents or slides requiring scanning. Further, the optimization may then assess a confidence threshold for the scanned image and determine if alterations are needed to fully capture the specifically intended information within the document or slide. Aspects of the present disclosure can be used to digitally capture the focal points of slides or documents during the scanning process, thereby mitigating the necessity to repeat digital scans. Referring now to FIG.17, an exemplary embodiment of an apparatus 1700 for inline scanned image enrichment is illustrated. Apparatus includes circuitry 1704 configured as described in more detail below. Circuitry 1704 may include a computing device. Computing device may include any computing device as described in this disclosure, including without limitation a microcontroller, microprocessor, digital signal processor (DSP) and/or system on a chip (SoC) as described in this disclosure. Computing device may be or include a configurable 61 Attorney Docket No.1519-029PCT1 hardware circuit. Computing device may include, be included in, and/or communicate with a mobile device such as a mobile telephone or smartphone. Computing device may include a single computing device operating independently, or may include two or more apparatus operating in concert, in parallel, sequentially or the like; two or more apparatus s may be included together in a single computing device or in two or more computing devices. Computing device may interface or communicate with one or more additional devices as described below in further detail via a network interface device. Network interface device may be utilized for connecting computing device to one or more of a variety of networks, and one or more devices. Examples of a network interface device include, but are not limited to, a network interface card (e.g., a mobile network interface card, a LAN card), a modem, and any combination thereof. Examples of a network include, but are not limited to, a wide area network (e.g., the Internet, an enterprise network), a local area network (e.g., a network associated with an office, a building, a campus or other relatively small geographic space), a telephone network, a data network associated with a telephone/voice provider (e.g., a mobile communications provider data and/or voice network), a direct connection between two computing devices, and any combinations thereof. A network may employ a wired and/or a wireless mode of communication. In general, any network topology may be used. Information (e.g., data, software etc.) may be communicated to and/or from a computer and/or a computing device. Computing device may include but is not limited to, for example, a computing device or cluster of computing devices in a first location and a second computing device or cluster of computing devices in a second location. Computing device may include one or more computing devices dedicated to data storage, security, distribution of traffic for load balancing, and the like. Computing device may distribute one or more computing tasks as described below across a plurality of computing devices of computing device, which may operate in parallel, in series, redundantly, or in any other manner used for distribution of tasks or memory between computing devices. Computing device may be implemented, as a non-limiting example, using a “shared nothing” architecture. Circuitry 1704 may alternatively or additionally include a Field Programmable Gate Array (FPGA), a hardware circuit with one or more memory elements that may contain biases, weights, coefficients, or other parameters, or the like. With continued reference to FIG.17, apparatus 1700 may be designed and/or configured to perform any method, method step, or sequence of method steps in any embodiment described 62 Attorney Docket No.1519-029PCT1 in this disclosure, in any order and with any degree of repetition. For instance, apparatus 1700 may be configured to perform a single step or sequence repeatedly until a desired or commanded outcome is achieved; repetition of a step or a sequence of steps may be performed iteratively and/or recursively using outputs of previous repetitions as inputs to subsequent repetitions, aggregating inputs and/or outputs of repetitions to produce an aggregate result, reduction or decrement of one or more variables such as global variables, and/or division of a larger processing task into a set of iteratively addressed smaller processing tasks. Apparatus 1700 may perform any step or sequence of steps as described in this disclosure in parallel, such as simultaneously and/or substantially simultaneously performing a step two or more times using two or more parallel threads, processor cores, or the like; division of tasks between parallel threads and/or processes may be performed according to any protocol suitable for division of tasks between iterations. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which steps, sequences of steps, processing tasks, and/or data may be subdivided, shared, or otherwise dealt with using iteration, recursion, and/or parallel processing. With continued reference to FIG.17, apparatus 1700 may receive a set of subject data 1708. As used in this disclosure, “subject data” is a set of information relating to a subject’s medical history. As a non-limiting embodiment, subject data 1708 may include, without limitation, any info supporting a diagnosis and treatment of a current or past ailment. Subject data 1708 may further include any of the following personal information: age, height, weight, heart rate, current diagnosis, medical history, allergies, current condition, current symptoms, known disorders, test results (e.g., basic metabolic panel, blood test, other physiological data, and the like), medications, growth chart, family history, medical treatment, and the like. Medical treatment may include medical evaluation, medical diagnosis, prescribing medicines/treatments, providing treatments, therapy, occupational therapy, physical therapy, care given during a hospital stay, surgery, treatment provided by a medical practitioner, and the like. Subject data 1708 may include information about a current condition such as a reported stomach pain of the patient or other medical history data. As used in the current disclosure, “medical history” is any data relating to the user’s medical care. Medical care may include any part of the attempt to improve the user’s health. Subject data 1708 may include past and present injuries or ailments suffered by the patient. Subject data 1708 may further include subject metadata including, name, 63 Attorney Docket No.1519-029PCT1 contact information, biometric data, location, and prior classifications whether based on symptoms, diagnoses, treatments, or any other data type but is in any way relevant for effectively capturing the targeted information. With continued reference to FIG.17, subject data 1708 may be manually entered using a remote display 1712. Remote display may incorporate a graphical user interface 1716 configured by circuitry 1704. For example, and without limitation, the user or a third party may manually input subject data 1708 using graphical user interface 1716 of circuitry 1704 and/or a remote display 1712, such as, for example, a smartphone or laptop. A “graphical user interface (GUI),” as used herein, is a graphical form of user interface that allows users to interact with electronic devices. In some embodiments, GUI 1716 may include icons, menus, other visual indicators, or representations (graphics), audio indicators such as primary notation, and display information and related user controls. A menu may contain a list of choices and may allow users to select one from them. A menu bar may be displayed horizontally across the screen such as pull-down menu. When any option is clicked in this menu, then the pull-down menu may appear. A menu may include a context menu that appears only when the user performs a specific action. An example of this is pressing the right mouse button. When this is done, a menu may appear under the cursor. Files, programs, web pages and the like may be represented using a small picture in a graphical user interface. For example, links to decentralized platforms as described in this disclosure may be incorporated using icons. Using an icon may be a fast way to open documents, run programs etc. because clicking on them yields instant access. Information contained in user interface may be directly influenced using graphical control elements such as widgets. A “widget,” as used herein, is a user control element that allows a user to control and change the appearance of elements in the user interface. In this context a widget may refer to a generic GUI element such as a check box, button, or scroll bar to an instance of that element, or to a customized collection of such elements used for a specific function or application (such as a dialog box for users to customize their computer screen appearances). User interface controls may include software components that a user interacts with through direct manipulation to read or edit information displayed through user interface. Widgets may be used to display lists of related items, navigate the system using links, tabs, and manipulate data using check boxes, radio boxes, and the like. In a non-limiting embodiment, remote display 1712 may refer to multiple 64 Attorney Docket No.1519-029PCT1 displays used collaterally to ingest, export, or share data across multiple platforms, as in a neural network described in detail below and in other figures. Still referring to FIG.17, in some embodiments, subject data 1708 may include handwritten, pixelated, or poorly printed documents previously scanned in but not machine- readable. These types of subject data 1708 may rely on optical character recognition or optical character reader (OCR), executed by circuitry 1704 to automatically convert images of written (e.g., typed, handwritten or printed text) into machine-encoded text. In some cases, recognition of at least a keyword from an image component may include one or more processes, including without limitation optical character recognition (OCR), optical word recognition, intelligent character recognition, intelligent word recognition, and the like. In some cases, OCR may recognize written text, one glyph or character at a time. In some cases, optical word recognition may recognize written text, one word at a time, for example, for languages that use a space as a word divider. In some cases, intelligent character recognition (ICR) may recognize written text one glyph or character at a time, for instance by employing machine learning processes. In some cases, intelligent word recognition (IWR) may recognize written text, one word at a time, for instance by employing machine learning processes. Still referring to FIG.17, in some cases OCR may be an "offline" process, which analyses a static document or image frame. In some cases, handwriting movement analysis can be used as input to handwriting recognition. For example, instead of merely using shapes of glyphs and words, this technique may capture motions, such as the order in which segments are drawn, the direction, and the pattern of putting the pen down and lifting it. This additional information can make handwriting recognition more accurate. In some cases, this technology may be referred to as “online” character recognition, dynamic character recognition, real-time character recognition, and intelligent character recognition. Still referring to FIG.17, in some cases, OCR processes may employ pre-processing of image component. Pre-processing process may include without limitation de-skew, de-speckle, binarization, line removal, layout analysis or “zoning,” line and word detection, script recognition, character isolation or “segmentation,” and normalization. In some cases, a de-skew process may include applying a transform (e.g., homography or affine transform) to image component to align text. In some cases, a de-speckle process may include removing positive and negative spots and/or smoothing edges. In some cases, a binarization process may include 65 Attorney Docket No.1519-029PCT1 converting an image from color or greyscale to black-and-white (i.e., a binary image). Binarization may be performed as a simple way of separating text (or any other desired image component) from a background of image component. In some cases, binarization may be required for example if an employed OCR algorithm only works on binary images. In some cases, a line removal process may include removal of non-glyph or non-character imagery (e.g., boxes and lines). In some cases, a layout analysis or “zoning” process may identify columns, paragraphs, captions, and the like as distinct blocks. In some cases, a line and word detection process may establish a baseline for word and character shapes and separate words, if necessary. In some cases, a script recognition process may, for example in multilingual documents, identify script allowing an appropriate OCR algorithm to be selected. In some cases, a character isolation or “segmentation” process may separate signal characters, for example character-based OCR algorithms. In some cases, a normalization process may normalize aspect ratio and/or scale of image component. Still referring to FIG.17, in some embodiments an OCR process will include an OCR algorithm. Exemplary OCR algorithms include matrix matching process and/or feature extraction processes. Matrix matching may involve comparing an image to a stored glyph on a pixel-by- pixel basis. In some case, matrix matching may also be known as “pattern matching,” “pattern recognition,” and/or “image correlation.” Matrix matching may rely on an input glyph being correctly isolated from the rest of the image component. Matrix matching may also rely on a stored glyph being in a similar font and at a same scale as input glyph. Matrix matching may work best with typewritten text. Still referring to FIG.17, in some embodiments, an OCR process may include a feature extraction process. In some cases, feature extraction may decompose a glyph into features. Exemplary non-limiting features may include corners, edges, lines, closed loops, line direction, line intersections, and the like. In some cases, feature extraction may reduce dimensionality of representation and may make the recognition process computationally more efficient. In some cases, extracted feature can be compared with an abstract vector-like representation of a character, which might reduce to one or more glyph prototypes. General techniques of feature detection in computer vision are applicable to this type of OCR. In some embodiments, machine- learning processes like nearest neighbor classifiers (e.g., k-nearest neighbors algorithm) can be used to compare image features with stored glyph features and choose a nearest match. OCR 66 Attorney Docket No.1519-029PCT1 may employ any machine-learning process described in this disclosure, for example machine- learning processes described with reference to other figures below. Exemplary non-limiting OCR software includes Cuneiform and Tesseract. Cuneiform is a multi-language, open-source optical character recognition system originally developed by Cognitive Technologies of Moscow, Russia. Tesseract is free OCR software originally developed by Hewlett-Packard of Palo Alto, California, United States. Still referring to FIG.17, in some cases, OCR may employ a two-pass approach to character recognition. Second pass may include adaptive recognition and use letter shapes recognized with high confidence on a first pass to recognize better remaining letters on the second pass. In some cases, two-pass approach may be advantageous for unusual fonts or low- quality image components where visual verbal content may be distorted. Another exemplary OCR software tool includes OCRopus. OCRopus development is led by German Research Centre for Artificial Intelligence in Kaiserslautern, Germany. In some cases, OCR software may employ neural networks, for example neural networks as taught in reference to other figures herein. Still referring to FIG.17, in some cases, OCR may include post-processing. For example, OCR accuracy can be increased, in some cases, if output is constrained by a lexicon. A lexicon may include a list or set of words that are allowed to occur in a document. In some cases, a lexicon may include, for instance, all the words in the English language, or a more technical lexicon for a specific field. In some cases, an output stream may be a plain text stream or file of characters. In some cases, an OCR process may preserve an original layout of visual verbal content. In some cases, near-neighbor analysis can make use of co-occurrence frequencies to correct errors, by noting that certain words are often seen together. For example, “Washington, D.C.” is generally far more common in English than “Washington DOC.” In some cases, an OCR process may make us of a priori knowledge of grammar for a language being recognized. For example, grammar rules may be used to help determine if a word is likely to be a verb or a noun. Distance conceptualization may be employed for recognition and classification. For example, a Levenshtein distance algorithm may be used in OCR post-processing to further optimize results. Still referring to FIG.17, subject data 1708 may additionally be generated via the answer to a series of questions. The series of questions may be implemented using a chatbot. A chatbot 67 Attorney Docket No.1519-029PCT1 may be configured to generate questions regarding the user’s current ailments, past ailments, medical history, family medical history, and the like. In a non-limiting embodiment, a user may be prompted to input specific information or may fill out a questionnaire. In an embodiment, GUI 1716 may display a series of questions to prompt a user for information pertaining to subject data 1708. As a further example, chatbot may display a list of possible conditions to user, from which user may select each applicable condition the user currently suffers from. In a non- limiting embodiment, circuitry 1704 may receive user selection of an autoimmune disorder and select additional autoimmune disorders as a function of the user selection. In another example, and without limitation, a medical professional may input subject data 1708 using GUI 1716 and/or remote display 1712. In another example, and without limitation, a third party, such as a different medical entity or healthcare provider, may transmit information of subject data 1708 to circuitry 1704. Subject data 1708 may be transmitted to circuitry 1704, such as via a wired or wireless communication, as previously discussed in this disclosure. Subject data 1708 may be retrieved from multiple sources including clinical reports, available medical facility records, or insurance databases to aid in the rendered medical service. Still referring to FIG.17, subject data 1708 may be received from an external data repository 1720. As used herein, “external data repository” is any accessible database containing information applicable to the specified subject and subject’s medical history but is not accessible within computing device’s 1704 internal network. External data repository may include hospital databases, insurance provider databases, emergency care databases, private practices, webcrawler searches, social media, or any other accessible medical information. If external data repository is a shared resource and available for uploads, completed scans from circuitry 1704 may be saved to external data repository 1720. In a non-limiting embodiment, where the subject’s birthday is not initially provided and not available within public records, circuitry 1704 may search social media posts related to subject to identify a best estimate of birth year and day. Circuitry 1704 may further assess a reliability score for each individual piece of subject data not directly provided by patient or user. In the case of the example above, where circuitry 1704 discovers a “Happy 30^th birthday!” post from April 21, 2016, circuitry 1704 may assume subject was 30 years old on the specified date and calculate a current age to autofill within subject data 1708. That age may then be assigned a low reliability score and flagged for update or validation at a later time. 68 Attorney Docket No.1519-029PCT1 Still referring to FIG.17, computing device receives subject data 1708 and uses it to reference the most applicable candidate set 1724 or build a new candidate set 1724. As used herein, “candidate set” is a plurality of models to inform the scanning process 1728 as well as detect features specific to the subject condition as derived from subject information. Candidate set 1724 may be determined or built by machine-learning processes. Once a best fit candidate set 1724 is declared by circuitry 1704, it is used to program scanning process 1728 characteristics to target the specific details sought based on the candidate set 1724 model. The declared candidate set 1724, which may be locally cached within circuitry 1704 and/or memory communicatively connected to circuitry 1704, such as memory electronically coupled to and/or combined in an electronic device with circuitry 1704, may then be either instantiated directly on the configured circuitry making up circuitry 1704, or stored in memory of or at the circuit from which it can be locally retrieved and instantiated. In a non-limiting embodiment, this instantiation of candidate set 1724 may use flash memory which retains the parameter sets for candidate models. In a separate non-limiting embodiment, circuitry 1704 may communicatively coordinate with external devices where candidate set 1724 is stored to send the information within subject data 1708 and any descriptors applied by circuitry 1704 to retrieve the appropriate candidate set for local use. In a non-limiting embodiment, a broken tibia candidate set would focus scanning process 1728 on x-ray scans with enhanced focus on fractures when x-ray document is labeled with patient’s name and may be labeled as tibia or related anatomical reference. This enhanced focus may adjust illumination, angles, focus, resolution, magnification, or other scan characteristics to narrowly target the fracture indications. Candidate set 1724 may be retrieved from an on-board memory 1736. On-board memory 1736 refers to any local repository where candidate sets 1724 may be saved and recalled without any external network access. Candidate set 1724 may further be retrieved from external data repository 1720, defined and described above. Candidate set 1724 may be built by circuitry 1704 when no available candidate set to sufficiently match subject data 1708 is available. In a non-limiting embodiment, in cases where no candidate set fits, circuitry 1704 may prompt user to specify the type(s) of data, location, key words, physical attributes, or other descriptive characteristic to focus the scanning process. When this type of additional manual input is necessary, circuitry 1704 may prompt user for additional feedback upon concluding scan. If user feedback affirms the results, circuitry 1704 may commit 69 Attorney Docket No.1519-029PCT1 the newly generated candidate set to either on-board memory 1736, or an available external data repository 1720 for future use. Still referring to FIG.17, scanning process 1728 conducts a scan of a plurality of documents or slides as assigned by user. Scanning process 1728 may be carried out by any equipment capable of converting tangible documents or images into digital documents or images. In a non-limiting embodiment, scanning process 1728 may be conducted by desktop scanners often referred to as flatbed scanners, sheetfed scanners wherein multiple documents may be preloaded, drum scanners for negatives or transparent film, portable scanners including handheld devices able to convert a photo to a digital scan, or any other type of scanning device. Candidate set 1724 programs both software and hardware adjustments to optimize the scan outputs from scanning process 1728. Scanning process 1728 may then execute multiple scans with slight variations in each such that a stack of scans may later be correlated within a continuum diffusion model. As used herein, a “continuum diffusion model” is a software and hardware package capable of aligning and aggregating multiple images such that they form a single polished, uninterrupted portrayal of the targeted subject data. Continuum diffusion model allows user to fully interrogate all of the captured information within a single viewing engagement rather than querying each individual image separately. Continuum diffusion model relies on software to detect similar features and the altered angle, magnification, focus, illumination, resolution or other scan characteristic adjustments made between each individual scan, such that model may compile, align, and correlate views in an aggregate form and maximize the advantages of each individual image. Each individual scan may be intentionally shifted by set amounts such that a downstream program may reassemble the stack of scans in a logical manner for ease of user viewing. This process may be accomplished purely in image analysis software, or it may rely on detecting the adjusted scan features and automatically diffuse the images together based on the scan source features. In a non-limiting embodiment, a stack of scans may enable user to toggle the viewing angle or depth by rotating the digital version, or zooming in and out as necessary through a tool capable of this type of model views. Scans, as used herein, may refer to a multitude of types of images to be digitally replicated. In a non-exhaustive, non-limiting embodiment, scans may include nuclear medicine scans, positron-emission tomography (PET) scans, Gallium-67 scans, results from photo-acoustic imagery, ultrasound technology, optical scanning techniques, Computerized Tomography (CT) scans, C-arm image-acquisitions, MRI 70 Attorney Docket No.1519-029PCT1 scans, or any other medical imaging use. In a separate, non-limiting embodiment, continuum diffusion model may enable the correlation and synthesis of a hematoxylin and eosin (H&E) stained slide with an Estrogen Receptor, Progesterone Receptor (ER/PR) test slide. Candidate set 1724 and scanning process 1728 may enable this type of fusion of differing slides when they relate to the same sample or type of material. In a separate, non-limiting embodiment, scanning process 1728 may correlate a tumor heterogeneity analysis across multiple slides by identifying correlated size, angle, color, quantity, lighting, depth, or other characteristics which suggest a sample is related to another sample. Still referring to FIG.17, scanning process 1728 may use candidate set 1724 to enrich the image inline with no additional obligatory user input. Scanning process 1728 may rely on machine-learning processes to evaluate and enhance the scanned image. With sufficient training data, scanning process 1728 may identify basic entities and specific arrangement of those basic entities, such as the specific nuclei types and order contained within the scan. Through the identified focus points delineated by candidate set 1724, scanning process 1728 may identify basic entities and their arrangement based on the concentration of nuclei, their arrangement relative to the other material in the scan, their size, shape, color, filaments or protrusions from the cell(s), dynamic changes or movements of the cell(s), or other recognizable visual characteristic(s). Examples of additional recognizable visual characteristics may include recognition of the physical characteristics of non-cell properties, such as surrounding blood vessel sizes, points of integration or other relational properties between the cell(s) and other matter, or the focal point material’s relative size, color, or shape as compared to other material within the scan. Scanning process 1728 may further identify the architecture of the basic entities. In a non-limiting embodiment, basic entity architecture may be indicative of a specific type of tissue or material (e.g. breast biopsy, ducts, muscle, nerves, blood vessels, etc.). When multiple scans are conducted based on the same patient data 1708, scanning process 1728 may recognize changes in the architecture or arrangement of the scanned material such that it may affirm or diagnose a condition based on the recognized physical characteristics in both scans (e.g. abnormal nuclei contained within a breast biopsy being indicative of a ductal carcinoma or tumor). Scanning process 1728 may also detect certain physical properties related to the scan. In a non-limiting embodiment, scanning process 1728 may detect the actual size of a tumor area or necrosis. This data may then be appended to the scan output, or displayed within remote display 71 Attorney Docket No.1519-029PCT1 1712 and/or GUI 1716. Training data and machine-learning processes are described in detail in reference to other figures below. Still referring to FIG.17, scanning process 1728 and candidate set 1724 are managed by machine-learning processes 1732. Machine-learning processes generally are described in detail below. Candidate set 1724 is assigned based on subject data 1708 and the details contained therein. Specifically, circuitry 1704 may rely on a subject data classifier to interpret, classify, and group the details of subject data 1708. In a non-limiting embodiment, subject data classifier may identify subject data 1708 containing multiple symptoms usually indicative of a pulmonary infection and assign a descriptor summarizing that characteristic based on the grouping of all affiliated symptoms. Once verified as appropriate either by direct user feedback, or by successful scans of images targeting the correct focal points as identified by a candidate set 1724 that is based on pulmonary infections, machine learning processes 1732 may then rely on this descriptor use for future analyses. With specific reference to machine-learning processes 1732, circuitry 1704 may rely on training data to improve candidate set 1724 assignment and scanning process 1728 focus and targeting processes. Training data improving these processes may be in the form of user feedback, or simulated scenarios wherein a set of simulated subject data is provided to circuitry 1704 and purposefully matched with a candidate set 1724 such that user may then provide feedback to either promote or suppress the match. In a non-limiting embodiment, user may compile subject data 1708 with symptoms obviously indicative of a brain tumor, but based on insufficient availability of candidate set 1724 to match the symptoms, circuitry 1704 recommends a candidate set 1724 for migraines. User may, through GUI 1716, suppress the match by correcting the proper candidate set 1724 to a more closely representative candidate set 1724 or by assembling a new model candidate set 1724 for brain tumors which may be used in the future for sets of subject data 1708 similarly describing a brain tumor. Machine-learning processes 1732 may conduct confidence checks of completed scans to ensure targeted data as identified by candidate set 1724 is sufficiently captured. Training data may further rely on user feedback to approve or disapprove of completed scans to train the confidence threshold applied. Still referring to FIG.17, scanning process 1728 may take additional steps beyond simply capturing a scan. Scanning process 1728 may also quantify the size and number of target matter, as in estimating the size of an identified tumor. Scanning process 1728, based on candidate set 1724, may identify certain conditions based on the physical properties (e.g. observed colors 72 Attorney Docket No.1519-029PCT1 following an H&E stain). As a separate example, after multiple runs of scanning process 1728, completed scan may still be inadequately capturing the targeted details, wherein machine- learning processes 1732 may identify, based on a confidence threshold assessment, the inadequacy. Machine-learning processes 1732 may then direct the scan to be reconducted, or conclude the document or slide is incompatible for data retrieval, wherein remote display 1712 and/or GUI 1716 would generate this feedback for user. This process of validating sufficient scan quality may depend on blur detection methods described in detail below in reference to other figures. In a non-limiting embodiment, confidence threshold may be strictly a user generated standard based on a plurality of instance of user feedback assessing a completed scan for satisfactory quality. Machine learning processes 1732 may rely on user feedback as training data to generate a lower threshold for scan quality, wherein any scan falling below the learned confidence threshold would immediately be re-captured to improve the quality. When insufficient details are contained within subject data 1708 to readily identify a fitting candidate set 1724, circuitry 1704 may initiate a chatbot communication, sourced either from circuitry 1704 or externally from an affiliated remote device in communication with circuitry 1704, with user to clarify the missing information. Chatbot engagement may rely on a decision tree or database to successfully guide the communication to acquiring the missing information. Still referring to FIG.17, scanning process 1728 may incorporate image processing algorithms to enhance the quality of the scanned product to enhance at least a viewability characteristic. These algorithms may enable resolution enhancements, light and shading adjustments, color optimization, smoothing, sharpening, or any other variation of image improvement. This processing step may be conducted prior to a confidence threshold assessment to improve the likelihood that the specifically intended data capture was sufficiently gathered. Image processing may also be completed subsequent to a satisfactory confidence threshold assessment, such that an image deemed good enough will still be improved prior to user viewing or storage. Scanning process 1728 may append certain pathology-related information deemed relevant and applicable to the completed scan. In a non-limiting embodiment, this type of appended information or user query may enable the user to direct apparatus 1700 to assess a tumor infiltrating lymphocyte within a tumor-associated stroma. Apparatus 1700, relying on machine-learning processes 1732, may identify and label the specific physical characteristics within the scanned slide supporting the assessment of the tumor infiltrating lymphocyte. 73 Attorney Docket No.1519-029PCT1 Additionally, user may be prompted with the option for further processing and may choose from a list of specific types of processing as presented through GUI 1716. Once a completed scan concludes any image processing operations, it is stored in an external data repository 1720, displayed using remote display 1712, or both. Additionally, each individual stack view may be saved either locally to circuitry 1704, to external data repository 1720, or both, such that they may be recalled upon user command or used as training data. Training data is discussed in detail below in reference to other figures. At a high level, aspects of the present disclosure are directed to a method for visualizing digitized slides comprising retrieving, by one or more computer processors, a digitized slide, determining, by the one or more computer processors, one or more visualization components of the digitized slide, generating, by the one or more computer processors, a virtual slide corresponding to the digitized slide based on the one or more visualization components, and displaying, by the one or more computer processors, a visualization of the virtual slide. In an embodiment, digitization of slides refers to the processing and synthetization of one or more slides in a manner capturing and aligning the information contained therein as well as combining multiple individual digitized slides into a single virtual slide as appropriate. Referring now to FIG.18, a block diagram illustrating an apparatus 1800 and method for visualization of digitized slides is shown. Apparatus 1800 may include a processor 1804, which is communicatively connected to and configured by a memory 1808. Processor 1804 may include a computer processor 1804. Apparatus 1800 may include any computing device as described in this disclosure, including without limitation a microcontroller, microprocessor, digital signal processor (DSP) and/or system on a chip (SoC) as described in this disclosure. apparatus 1800 may include, be included in, and/or communicate with a mobile device such as a mobile telephone or smartphone. Apparatus 1800 may include a single computing device operating independently, or may include two or more computing device operating in concert, in parallel, sequentially or the like; two or more computing devices may be included together in a single computing device or in two or more computing devices. Apparatus 1800 may interface or communicate with one or more additional devices as described below in further detail via a network interface device. Network interface device may be utilized for connecting apparatus 1800 to one or more of a variety of networks, and one or more devices. Examples of a network interface device include, but are not limited to, a network interface card (e.g., a mobile network 74 Attorney Docket No.1519-029PCT1 interface card, a LAN card), a modem, and any combination thereof. Examples of a network include, but are not limited to, a wide area network (e.g., the Internet, an enterprise network), a local area network (e.g., a network associated with an office, a building, a campus or other relatively small geographic space), a telephone network, a data network associated with a telephone/voice provider (e.g., a mobile communications provider data and/or voice network), a direct connection between two computing devices, and any combinations thereof. A network may employ a wired and/or a wireless mode of communication. In general, any network topology may be used. Information (e.g., data, software etc.) may be communicated to and/or from a computer and/or a computing device. Apparatus 1800 may include but is not limited to, for example, a computing device or cluster of computing devices in a first location and a second computing device or cluster of computing devices in a second location. Apparatus 1800 may include one or more computing devices dedicated to data storage, security, distribution of traffic for load balancing, and the like. Apparatus 1800 may distribute one or more computing tasks as described below across a plurality of computing devices of computing device, which may operate in parallel, in series, redundantly, or in any other manner used for distribution of tasks or memory between computing devices. Apparatus 1800 may be implemented, as a non-limiting example, using a “shared nothing” architecture. With continued reference to FIG.18, processor 1804 may be designed and/or configured to perform any method, method step, or sequence of method steps in any embodiment described in this disclosure, in any order and with any degree of repetition. Processor 1804 may be configured to perform instructions, including steps, encoded in memory 1808. For instance, processor 1804 may be configured to perform a single step or sequence repeatedly until a desired or commanded outcome is achieved; repetition of a step or a sequence of steps may be performed iteratively and/or recursively using outputs of previous repetitions as inputs to subsequent repetitions, aggregating inputs and/or outputs of repetitions to produce an aggregate result, reduction or decrement of one or more variables such as global variables, and/or division of a larger processing task into a set of iteratively addressed smaller processing tasks. Processor 1804 may perform any step or sequence of steps as described in this disclosure in parallel, such as simultaneously and/or substantially simultaneously performing a step two or more times using two or more parallel threads, processor cores, or the like; division of tasks between parallel threads and/or processes may be performed according to any protocol suitable for division of 75 Attorney Docket No.1519-029PCT1 tasks between iterations. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which steps, sequences of steps, processing tasks, and/or data may be subdivided, shared, or otherwise dealt with using iteration, recursion, and/or parallel processing. With continued reference to FIG.18, processor 1804 may retrieve a digitized slide 1812a–c. As used in this disclosure, a "digitized slide" is a digital image that represents at least a portion of a slide or at least a section of tissue. In some cases, a digitized slide 1812a–c may include a representation of a section of tissue 1813a (i.e., histological section). Alternatively or additionally, the digitized slide 1812a–c may include a plurality of tissue sections 1813a–c; the plurality of tissue sections may be sectioned from spatially adjacent tissue, for instance within a tissue block 1814. As used in this disclosure, a "tissue section" is a slice of material, which allows for two-dimensional imaging of the material. In some cases a tissue section may include a slice of a biological material, such as human soft tissue. For example, tissue sections 1813a–c may be considered as two-dimensional representations of tissue, e.g., in an X-Y plane, and adjacent tissue sections may be considered two-dimensional slices at different but adjacent locations along a third dimension, e.g., Z-axis. As used in this disclosure, "tissue block" may include a volume of material. In some cases, a tissue block may be considered in three- dimensions, where a tissue section may be considered only in two dimensions. In some cases, tissue sections may be sliced from a tissue block. In some cases, a tissue block may include a biological material. For instance, tissue block may include human soft tissue from a biopsy. In some cases, a tissue block is solid; alternatively or additionally a tissue block may include a fluid. Still referring to FIG.18, a digitized slide may include a digital representation of a glass slide on which biomedical specimens are mounted to digital images of the slide usually through the use of a scanning process. The images may be associated with or include metadata indicating, for example, the location of a pathology specimen on the slide as well as information about presence and location of components like annotation, bubbles, and debris. As used in this disclosure, "metadata" refers to information about information, for instance information about digitized slide, virtual slide, tissue section, and/or tissue block. Metadata may include information about tissue, patient, slides, or the like. As described above, retrieving the digitized slide may be accomplished through a network connected communication, connecting a digital 76 Attorney Docket No.1519-029PCT1 storage device containing digitized slides, directly capturing the digitized slide through local scanning operations, or any other mechanism to convey digitized slides. Still referring to FIG.18, in some embodiments, processor 1804 may determine that digitized slide 1812a is associated with at least a member of a set of digitized slides 1812a–c. In some cases processor 1804 may determine, based on metadata associated with digitized slide 1812a, that the digitized slide 1812a is a member of a set of digitized slides 1812a–c associated with at least one of a patient case 1815 or a tissue block 1814. As used in this disclosure, a "patient case" refers to information associated with a patient. For instance, patient case may include a diagnosis, prognosis, or other medical record or health data. Patient case may also include demographic information. With continued reference to FIG.18, processor 1804 may determine at least a visualization component 1816 of digitized slide 1812a–c. As used herein, “visualization component” refers to a categorizable characteristic of a digitized slide, such as tissue section, associated metadata individual artifact, or an annotation contained in the initial digitized slide(s). A visualization component assessment module may be used to evaluate and process the visualization components. This visualization component 1816 assessment may use a machine- learning process 1820 to analyze and determine visualization components. In a non-limiting embodiment, processor 1804 may automatedly identify commonalities between digitized slides, such as the same source patient, same biopsy material type, same timing, or similarities between any other available metadata. Alternatively or additionally, processor 1804 may purposefully combine non-common visualization components in cases where highlighting the difference may be illustrative of the targeted visualization. Still referring to FIG.18, in some embodiments, at least a visualization component 1816 may include a representation of at least one of a tissue section, an artifact, or an annotation. As used in this disclosure, an "artifact" is an element of an image, or object being imaged, which is not naturally or organically present. For instance, an artifact may include an image of a glass slide (or materials on the glass slide) rather than the tissue slide with the glass slide. Processor 1804 may determine, by the one or more computer processors, one or more visualization components of the digitized slide. Still referring to FIG.18, processor 1804 may generate a virtual slide 1824a corresponding to digitized slide 1812a. In some cases, processor 1804 may generate virtual slide 77 Attorney Docket No.1519-029PCT1 1824a based on at least a visualization component 1816. As used in this disclosure, a "virtual slide" is a digital slide that represents at least a portion of a slide or at least a section of tissue and is at least partially virtual. For instance, a virtual slide may include virtual imagery. Virtual imagery may include portions of a digital slide, for instance to correct debris, artifacts, and/or annotations. Virtual imagery may include an entire virtual slide or digital representation of a tissue section, for instance where imagery representing an intra-serial section slide tissue is transformed through a registration transform. Views of digitized slides can be customized to filter out components like annotations, debris and bubbles. In some embodiments, virtual slides can be created by repositioning (e.g., transforming) the pathology specimen in images to enable ergonomic viewing. This repositioning may be accomplished using affine transformation, homography transform, or the like. Using affine transformation may enable conversion of angles between lines or distances between points, and preserve relative ratios of distances between points lying on a straight line. Affine transformations may be accomplished through an affine transformation matrix, wherein an augmented vector and matrix are used to represent the translation and the linear map of the digitized slide. Then using matrix multiplication, the affine transformation matrix is converted by multiplying the initial finite-dimensional representation by an invertible matrix A, and the translation occurs through the addition of a vector b to create the final, affine transformed matrix, y, as shown below: ^^ ൌ ^^^ ^^^ ൌ ^^ ^^ ^ ^^ Affine transformations intentionally preserve collinearity between points, parallelism between lines, convexity of sets, and ratios of lengths while altering the orientation or size, or both. As an example, and still referring to FIG.18, simple two-dimensional translational transformations may be described using a vector (V) with two components Vx, Vy that describes displacement of blocks and/or pixels in an image. More complex transformations such as rotation, zooming, and warping may be described using affine transformations. Some exemplary affine transformations use four-parameter or six-parameter affine models. For example, a six-parameter affine transformation may be described as: x’ = ax + by + c y’ = dx + ey + f 78 Attorney Docket No.1519-029PCT1 A four-parameter affine transformation may be described as: x’ = ax + by + c y’ = -bx + ay + f where (x,y) and (x’,y’) are pixel locations in before and after transformation, respectively; a, b, c, d, e, and f are parameters of the affine motion model. Still referring to FIG.18, processor 1804 may display a visualization 1825 of virtual slide 1824a. As used herein, “visualization,” where used as a noun, refers to a displayed representation. As used herein, “visualization,” where used as a verb, is the process of representing and displaying. For instance, visualization 1825 of a virtual slide 1824a may include presentation of a digitized image (e.g., digitized slide and/or virtual slide) on a display. Virtual slide 1824 generator module may be used to conduct these modifications. In a non-limiting embodiment, machine-learning processes 1820 may identify a series of discrete slide scans, which have a common characteristic across the slides, such as being from the same patient, same tissue type, or same visible disorder. Virtual slide 1824 generator may then rely on machine- learning processes 1820 to combine those slides in a grid format such that the user is able to view all of the common characteristics in the slides within a single view. Special options can be enabled for tissue slides that are part of the same patient case. In case the slides are for the same patient and are derived from the same tissue block, these slides are referred to as inter-serial section slides. The specific order and layout may be modified by user selection, machine- learning training data, or a default numeric or chronological organization method. Training data supporting the display format may be sourced from prior user engagements where a certain layout was affirmed as effective by a user. Training data and machine-learning generally is covered in detail in reference to other figures herein. In some cases, processor 1804 may display a plurality of virtual slides 1824a–c, including, for instance, a virtual slide 1824a, corresponding to a set of digitized slides 1812a–c. Still referring to FIG.18, processor 1804 may be configured to determine at least a user- configurable option 1826 associated with virtual slide 1824a-c based on at least a visualization component 1816. As used in this disclosure, a "user-configurable option" is a parameter which user may control. For instance, in some cases, user-configurable options may include parameters associated with at least a visualization component, digitized slide, virtual slide, whole slide image viewer, or the like. In some cases, at least a user-configurable option 1826 may be 79 Attorney Docket No.1519-029PCT1 determined by accessing a look-up table 1827. As used in this disclosure, a "look-up table" is a corpus of indexable data. In some cases, a look-up table may include data organized in a table; alternatively or additionally a look-up table may include a database or any other data structure. In some versions, look-up table 1827 may be indexed by at least a visualization component 1816. In some cases, processor 1804 may be configured to display visualization 1825 using a user display 1828. In some cases, processor 1804 may be configured to display visualization via a whole slide image viewer 1829. As used in this disclosure, a "whole slide image viewer" is a system, apparatus, and/or module configured to display a whole slide image. In some cases, a whole slide image viewer may include software for viewing a digital image of a slide and/or tissue section. In some cases, a whole slide image viewer may be configured to display a virtual slide. An exemplary whole slide image viewer includes QuPath, open-source software available at qupath.github.io. In some versions, at least a user-configurable option 1826 may be presented to a user via a user display 1828 and/or whole slide image viewer 1829. In some cases, at least a user-configurable option 1826 may be entered, modified, selected, or the like using a user interface 1830. User interface may include any user interface described in this disclosure, including without limitation keyboard, mouse, other peripherals, as well as remote devices, such as smart phones, tablets, remote computing devices, and the like. Still referring to FIG.18, in some embodiments, processor 1804 may be further configured to receive a request to customize visualization 1825. In some embodiments, processor 1804 may be further configured to receive a request to display a second visualization 1825 of a different virtual slide 1824b. Second virtual slide 1824b may include any virtual slide described in this disclosure. In some embodiments, processor 1804 may be configured to determine a recommended set of visualization components 1816 to include in visualization 1825. In some cases, processor 1804 may be configured to determine a revised set of visualization components 1816 to include in visualization 1825 based on a user selection. A user selection may include a user configured option 1826 or any other user input, for instance from user interface 1830. Still referring to FIG.18, in some embodiments, processor 1804 may be configured to determine that digitized slide 1812a–c corresponds to an intra-serial section slide. As used in this disclosure, an "intra-serial section slide" is a slide with multiple serial sections mounted on it. In some cases, processor 1804 may determine correspondence between digitized slide 1812a–c and intra-serial section slide based on a presence of a plurality of serial sections 1813a–b in digitized 80 Attorney Docket No.1519-029PCT1 slide. As used in this disclosure, "serial sections" are tissue sections from a single tissue block. In some cases, serial sections may include tissue sections from adjacent locations within tissue block. Alternatively or additionally, serial sections may include tissue sections taken from locations within tissue block, separated by some depth, e.g., 0.01mm, 0.02mm, 0.05mm, 0.1mm, or the like. In some cases, finding a correspondence between digitized slide 1812a–c and intra- section serial slide may include classifying, by processor 1804, plurality of serial sections into a reference serial section and at least a remaining serial section and aligning, by processor 1804, the at least a remaining serial section to the reference serial section, yielding a plurality of aligned serial sections. In some versions, visualization 1825 of virtual slide 1824a–c may include plurality of aligned serial sections. In some cases, at least a remaining serial section may be aligned with reference serial section by computing, independently for each of the at least a remaining serial section, at least a registration transform relative to the reference serial section. As used in this disclosure, a "registration transform" is a transform, e.g., affine transform, which registers a digital image to a reference digital image. For instance, in some cases, a registration transform may re-orient data within a digitized intra-serial section slide, so that imagery representing a first serial section is oriented consistent with imagery representing a reference serial section. Registration transform may include any transform described in this disclosure. In some cases, plurality of aligned serial sections may be displayed (e.g., with user display 1828) in same order that corresponding plurality of serial sections appear on digital slide. In some cases, plurality of aligned serial sections may be spatially arranged within visualization 1825 based on a user-selected configuration 1826. In some versions, plurality of aligned serial sections may be spatially arranged in a compact representation such that the plurality of aligned serial sections appear closer to one another in visualization 1825 than in digitized slide 1812a–c. In some cases, at least a visualization component 1816 may include at least one annotation. In some versions, at least one annotation may be included in visualization 1825, for instance based on a user- configurable filter 1826. In some cases, aligning at least a remaining serial section to reference serial section may include aligning at least one annotation to the reference serial section. Still referring to FIG.18, in some cases, at least a registration transform may be computed based on a macro image 1831, for instance of digitized slide 1812a–c. As used in this disclosure, a "macro image" is a representation of an object, e.g., slide, which is at a relatively lower magnification than a non-macro image. For instance, in some cases, a macro image may be 81 Attorney Docket No.1519-029PCT1 generated using a macro lens with a lower magnification that permits optical imaging of a significant portion of a whole slide, e.g., whole slide. Macro image may be acquired using a macro camera. In some versions, macro camera and/or macro image 1831 may have a field of view that covers entire digitized slide 1812a–c and/or each serial section of plurality of serial sections. Still referring to FIG.18, in some embodiments, processor 1804 may be configured to store at least a registration transform in a non-volatile storage medium. Processor 1804 may be configured to acquire, through any means described in this disclosure (e.g., camera, sensor, data communication, network of the like), a whole slide image (WSI) 1832. As used in this disclosure, a "whole slide image" is an image that represents at least half of an entire slide. For instance, in some cases a whole slide image 1832 may represent an entire slide. In some cases, a whole slide image 1832 may include a macro image 1831. In some cases, a whole slide image 1832 may be generated through aggregating (e.g., stitching) a plurality of digitized slide 1812a–c images. Typically, a whole slide image 1832 may have a higher magnification than macro image 1831. Processor 1804 may be configured to compute based on at least a stored registration transform, at least a corresponding high-magnification registration transform applicable to WSI 1832. Processor 1804 may be configured to apply at least a high magnification registration transform to plurality of serial sections within WSI 1832 to yield a virtual WSI having a plurality of aligned serial sections. As used in this disclosure, a "virtual whole slide image" is a digital slide that represents a whole slide image and is at least partially virtual. For instance, a virtual whole slide image may include virtual imagery. Virtual imagery may include portions of a digital slide, for instance to correct debris, artifacts, and/or annotations. Virtual imagery may include an entire virtual whole slide image or digital representation of a tissue section, for instance where all representations of all serial sections on a slide experience registration transforms. In some cases, displaying visualization 1825 of virtual slide may include displaying a visualization 1825 of virtual WSI. Still referring to FIG.18, a slide with multiple serial sections mounted on it may be referred to as an intra-serial section slide. Serial sections of tissue are drawn from the same tissue block and contain consecutive or nearby slices of tissue, such that the tissue samples are expected to be of the same (or similar) shape as long as the mounting process on the glass slide does not significantly deform the serial sections. The visualization techniques presented herein 82 Attorney Docket No.1519-029PCT1 may facilitate pathological assessment of the tissue on intra-serial sections slides and make the evaluation of such slides less taxing. In some embodiments, processor 1804 may create the virtual slide with respect to placement of re-oriented serial sections. In some embodiments, alignment capabilities may be provided, e.g., portions of a virtual slide corresponding to an intra- serial section slide may be aligned to one another such that serial sections are displayed with the same orientation. Still referring to FIG.18, processor 1804 may enable magnification modifications within virtual slide 1824. As used herein, “magnification modification” refers to adjustments made to magnification. For instance magnification modification may include an adjustment to relative size of digital slide image while retaining all proportionality and parallelistic relationships of the digital slide image. Generally, the magnification may refer to the ability to zoom in or out on a specified digital slide image to improve user viewership of the identified key characteristics. The implementation of these magnification modifications may be accomplished by the virtual slide 1824 generator module relying machine-learning processes 1820 to identify and implement the appropriate transformation algorithm. These magnification modifications may be executed independent of any orientation modifications, or combined in a manner to align orientation and sizing concurrently. Consistent with such embodiments, one or more transforms used for reorientation of serial sections with respect to the reference serial section may be computed using a low magnification image (e.g., a macro image that has the glass slide in a single field of view of the macro camera). The transforms may then be adapted from the low magnification to be used with higher magnification images at which a larger resolution (e.g., in gigapixels) whole slide image (WSI) is captured. In this manner, a macro virtual image (e.g., low resolution) and WSI virtual image (e.g., high resolution) may be created. Still referring to FIG.18, processor 1804 may enable the filtration of noise or other undesirable components prior to generating a virtual slide. For example, machine-learning processes 1820 or the user may be able to identify and filter components like annotations, bubbles, and debris during creation of the virtual slide. Filtration may be accomplished by various adaptive processing methods to reduce or eliminate noise within the digital slide images. In a non-limiting embodiment, adaptive processing may rely on least squares, column, Wiener or Kalman filters to execute the noise filtering processes. 83 Attorney Docket No.1519-029PCT1 Still referring to FIG.18, processor 1804 may use machine-learning processes 1820 to identify and reduce or remove distortion within the digital slide images. In a non-limiting embodiment, the distortion removal process may convert any lenticular distortion by modifying the identified bending through digital transformation mechanisms. Processor 1804 may additionally enable occlusion correction where desired information is blocked or lost. In a non- limiting embodiment, processor 1804 may rely on machine-learning processes 1820 to perform generative interpolation Training data supporting distortion removal and occlusion correction machine-learning processes may be ingested from any communicatively connected machine- learning device with distortion removal historical data, especially within the field of digitally scanned images. Still referring to FIG.18, processor 1804 may implement one or more aspects of “generative artificial intelligence (AI),” a type of AI that uses machine-learning algorithms to create, establish, or otherwise generate data such as, without limitation, virtual slides and/or the like in any data structure as described herein (e.g., text, image, video, among others) that is similar to one or more provided training examples. In an embodiment, machine-learning processes 1820 may generate one or more generative machine-learning models that are trained on one or more sets of historical virtual slide generations. One or more generative machine- learning models may be configured to generate new examples that are similar to the training data of the one or more generative machine-learning models but are not exact replicas; for instance, and without limitation, data quality or attributes of the generated examples may bear a resemblance to the training data provided to one or more generative machine-learning models, wherein the resemblance may pertain to underlying patterns, features, or structures found within the provided training data. Still referring to FIG.18, in some cases, generative machine-learning models may include one or more generative models. As described herein, “generative models” refers to statistical models of the joint probability distribution ^^^ ^^, ^^^ on a given observable variable x, representing features or data that can be directly measured or observed (e.g., scanned digital slide) and target variable y, representing the outcomes or labels that one or more generative models aims to predict or generate (e.g., image noise distribution). In some cases, generative models may rely on Bayes theorem to find joint probability; for instance, and without limitation, Naïve Bayes classifiers may be employed by processor 1804 to categorize input data such as, 84 Attorney Docket No.1519-029PCT1 without limitation, scanned slide images into different classes such as, without limitation, scanned slide images that are very noisy, slightly discolored, or visually coherent. In a non-limiting example, and still referring to FIG.18, one or more generative machine- learning models may include one or more Naïve Bayes classifiers generated, by processor 1804, using a Naïve bayes classification algorithm. Naïve Bayes classification algorithm generates classifiers by assigning class labels to problem instances, represented as vectors of element values. Class labels are drawn from a finite set. Naïve Bayes classification algorithm may include generating a family of algorithms that assume that the value of a particular element is independent of the value of any other element, given a class variable. Naïve Bayes classification algorithm may be based on Bayes Theorem expressed as P(A/B)= P(B/A) P(A)÷P(B), where P(A/B) is the probability of hypothesis A given data B also known as posterior probability; P(B/A) is the probability of data B given that the hypothesis A was true; P(A) is the probability of hypothesis A being true regardless of data also known as prior probability of A; and P(B) is the probability of the data regardless of the hypothesis. A naïve Bayes algorithm may be generated by first transforming training data into a frequency table. Processor 1804 may then calculate a likelihood table by calculating probabilities of different data entries and classification labels. Processor 1804 may utilize a naïve Bayes equation to calculate a posterior probability for each class. A class containing the highest posterior probability is the outcome of prediction. Still referring to FIG.18, although Naïve Bayes classifier may be primarily known as a probabilistic classification algorithm; however, it may also be considered a generative model described herein due to its capability of modeling the joint probability distribution ^^^ ^^, ^^^ over observable variables X and target variable Y. In an embodiment, Naïve Bayes classifier may be configured to make an assumption that the features X are conditionally independent given class label Y, allowing generative model to estimate the joint distribution as ^^^ ^^, ^^^ ൌ ^^^ ^^^∏ ^^ ^^^ ^^ ^^ ∣ ^^^, wherein ^^^ ^^^ may be the prior probability of the class, and ^^^ ^^_^| ^^^ is the conditional probability of each feature given the class. One or more generative machine-learning models containing Naïve Bayes classifiers may be trained on labeled training data, estimating conditional probabilities ^^^ ^^_^| ^^^ and prior probabilities ^^^ ^^^ for each class; for instance, and without limitation, using techniques such as Maximum Likelihood Estimation (MLE). One or more generative machine-learning models containing Naïve Bayes classifiers may select a class label ^^ according to prior distribution ^^^ ^^^, and for each feature ^^_^, sample at least a value 85 Attorney Docket No.1519-029PCT1 according to conditional distribution ^^^ ^^_^| ^^^. Sampled feature values may then be combined to form one or more new data instance with selected class label ^^. In a non-limiting example, one or more generative machine-learning models may include one or more Naïve Bayes classifiers to generate new examples of virtual slides based on visually coherent historical virtual slides, wherein the models may be trained using training data containing a plurality of features e.g., features of user-identified exemplary visually coherent virtual slides, and/or the like as input correlated to a plurality of labeled classes e.g., very noisy virtual slides corrected to a visually coherent level as output. Still referring to FIG.18, in some cases, one or more generative machine-learning models may include generative adversarial network (GAN). As used in this disclosure, a “generative adversarial network” is a type of artificial neural network with at least two sub models (e.g., neural networks), a generator, and a discriminator, that compete against each other in a process that ultimately results in the generator learning to generate new data samples, wherein the “generator” is a component of the GAN that learns to create hypothetical data by incorporating feedbacks from the “discriminator” configured to distinguish real data from the hypothetical data. In some cases, generator may learn to make discriminator classify its output as real. In an embodiment, discriminator may include a supervised machine-learning model while generator may include an unsupervised machine-learning model as described in further detail with reference to other figures. Still referring to FIG.18, processor 1804 may enable combining one or more digitized slide images to create an enhanced image. Using machine-learning processes 1820, processor 1804 may identify images of the same sample using common metadata and/or unique image characteristics, then combine the images to create a more illustrative virtual slide. Various techniques are available for this form of image stacking. In a non-limiting embodiment, processor 1804 may use an extended depth of field (EDoF) technique to determine the size of the collected image stack to improve the focus of the final virtual slide. Processor 1804 may additionally increase contrast within the digitized slide(s) by classifying each pixel in the image based on contrast levels. After classifying, processor 1804 may then conduct edge finding operations to identify borders and enhance the image by making the borders crisper and any enclosed shapes clearer. In some implementations, contrast improvement may be performed using interpolation filters such as sub-pixel prediction filters. Interpolation filters may include, as 86 Attorney Docket No.1519-029PCT1 a non-limiting example, any filters described above, a low-pass filter, which may be used, without limitation, by way of an up-sampling process whereby pixels between pixels of block and/or frame previous to scaling may be initialized to zero, and then populated with an output of the low-pass filter. Alternatively or additionally, any luma sample interpolation filtering process may be used. Luma sample interpretation may include computation of an interpolated value at a half-sample interpolation filter index, falling between two consecutive sample values of a non- scaled sample array. Computation of interpolated value may be performed, without limitation, by retrieval of coefficients and/or weights from lookup tables; selection of lookup tables may be performed as a function of motion models of coding units and/or scaling ratio amounts, for instance as determined using scaling constants as described above. Computation may include, without limitation, performing weighted sums of adjacent pixel values, where weights are retrieved from lookup tables. Computed values may alternatively or additionally be shifted. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various alternative or additional implementations that may be used for interpolation filters. With continued reference to FIG.18, in an embodiment, discriminator may include one or more discriminative models, i.e., models of conditional probability ^^^ ^^| ^^ ൌ ^^^ of target variable Y, given observed variable X. In an embodiment, discriminative models may learn boundaries between classes or labels in given training data. In a non-limiting example, discriminator may include one or more classifiers as described in further detail below with reference to FIG.2 to distinguish between different categories e.g., blurry vs. clear, or states e.g., TRUE vs. FALSE within the context of generated data such as, without limitations, generated virtual slides, and/or the like. In some cases, computing device may implement one or more classification algorithms such as, without limitation, Support Vector Machines (SVM), Logistic Regression, Decision Trees, and/or the like to define decision boundaries. In a non-limiting example, and still referring to FIG.18, generator of GAN may be responsible for creating synthetic data that resembles real generated virtual slides. In some cases, GAN may be configured to receive scanned digital slides such as, without limitation, images of a kidney biopsy, as input and generates corresponding virtual slides aligning the kidney biopsy containing information describing or evaluating the performance of one or more image clarity aspects, especially in regard to a disease only perceivable in a visually coherent clarity. On the other hand, discriminator of GAN may evaluate the authenticity of the generated content by 87 Attorney Docket No.1519-029PCT1 comparing it to a historical, user-assessed successful kidney biopsy, for example, discriminator may distinguish between genuine and generated content and providing feedback to generator to improve the model performance. With continued reference to FIG.18, in other embodiments, one or more generative models may also include a variational autoencoder (VAE). As used in this disclosure, a “variational autoencoder” is an autoencoder (i.e., an artificial neural network architecture) whose encoding distribution is regularized during the model training process in order to ensure that its latent space includes desired properties allowing new data sample generation. In an embodiment, VAE may include a prior and noise distribution respectively, trained using expectation- maximization meta-algorithms such as, without limitation, probabilistic PCA, sparse coding, among others. In a non-limiting example, VEA may use a neural network as an amortized approach to jointly optimize across input data and output a plurality of parameters for corresponding variational distribution as it maps from a known input space to a low-dimensional latent space. Additionally, or alternatively, VAE may include a second neural network, for example, and without limitation, a decoder, wherein the “decoder” is configured to map from the latent space to the input space. In a non-limiting example, and still referring to FIG.18, VAE may be used by processor 1804 to model complex relationships between various types of digital scans or affiliated metadata. In some cases, VAE may encode input data into a latent space, capturing a visually clear virtual slide. Such encoding process may include learning one or more probabilistic mappings from observed digital slide scans to a lower-dimensional latent representation. Latent representation may then be decoded back into the original data space, therefore reconstructing the digital slide scan. In some cases, such decoding process may allow VAE to generate new examples or variations that are consistent with the learned distributions, including improved clarity digital slide scan outputs. Still referring to FIG.18, processor 1804 may configure generative machine-learning models to analyze input data such as, without limitation, noisy digital slide scans to one or more predefined templates such as visually clear, user-promoted virtual slides representing correct virtual slide format and clarity as described above, thereby allowing processor 1804 to identify discrepancies or deviations from virtual slide layouts, formats, focal points, or other viewability characteristics. In some cases, processor 1804 may be configured to pinpoint specific errors in 88 Attorney Docket No.1519-029PCT1 the received digitized slide 1812. In a non-limiting example, processor 1804 may be configured to implement generative machine-learning models to incorporate additional models to align and assemble virtual slides containing multiple digital slide sources as inputs. In some cases, errors may be classified into different categories or severity levels. In a non-limiting example, some errors may be considered minor, and generative machine-learning model such as, without limitation, GAN may be configured to generate virtual slides containing only slight adjustments while others may be more significant and demand more substantial corrections. In some embodiments, processor 1804 may be configured to flag or highlight blurred or distorted virtual slide images, altering the level of magnification and/or noise filtration implemented to deliver, directly on the input digitized slide 1812 using one or more generative machine-learning models described herein. In some cases, one or more generative machine-learning models may be configured to generate and output indicators such as visual indicators and/or any other indicators as described above. Such indicators may be used to signal the detected error described herein. Still referring to FIG.18, in some cases, processor 1804 may be configured to identify and rank detected common deficiencies (e.g., blurriness, glare, discoloration, positional discrepancy, orientation, etc.) across a plurality of digitized slide 1812 storage locations. Such ranking process may enable a prioritization of most prevalent issues, allowing users or processor 1804 to address the virtual slide display issues. In a non-limiting example, a detected glare that only appears in a location of the digitized slide 1812 that does not interfere with any biopsy analysis may be ranked as low importance as compared to a detected foggy or indistinct biopsy image caused by image pixelation and precluding an effective user analysis of the slide. Still referring to FIG.18, in some cases, one or more generative machine-learning models may also be applied by processor 1804 to edit, modify, or otherwise manipulate existing data or data structures. In an embodiment, output of training data used to train one or more generative machine-learning models such as GAN as described herein may include corrective modifications in prior digitized slide 1812 analyses that were declared as correct and effective by the prior user(s) which visually demonstrate modified digitized slide 1812 e.g., interpolating image data to improve clarity, and/or the like. In some cases, certain types of virtual slides may be synchronized with specified digitized slides 1812, for example, and without limitation, where a certain user prefers a specified layout and grouping method for a specified biopsy type, processor 1804 may automatedly bias the digitized slides 1812 that match the designated biopsy type. 89 Attorney Docket No.1519-029PCT1 With continued reference to FIG.18, other exemplary embodiments of generative machine-learning models may include, without limitation, long short-term memory networks (LSTMs), (generative pre-trained) transformer (GPT) models, mixture density networks (MDN), and/or the like. An ordinary person skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various generative machine-learning models that may be used to generate the formatted virtual slide output based on the retrieved digitized slide 1812 inputs. Still referring to FIG.18, in a further non-limiting embodiment, machine-learning processes 1820 may be further configured to generate a multi-model neural network that combines various neural network architectures described herein. In some cases, multi-model neural network may also include a hierarchical multi-model neural network, wherein the hierarchical multi-model neural network may involve a plurality of layers of integration; for instance, and without limitation, different models may be combined at various stages of the network. Convolutional neural network (CNN) may be used for image feature extraction, followed by LSTMs for sequential pattern recognition, and a MDN at the end for probabilistic modeling. Other exemplary embodiments of multi-model neural network may include, without limitation, ensemble-based multi-model neural network, cross-modal fusion, adaptive multi- model network, among others. An ordinary person skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various generative machine-learning models that may be used to modify the digitized slides 1812 described herein. An ordinary person skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various multi-model neural networks and combinations thereof that may be implemented by processor 1804 consistent with this disclosure. Still referring to FIG.18, where processor 1804 detects that the digitized slide 1812 may contain some amount of blurriness, processor 1804 may implement one or more corrective mechanisms to improve clarity. In a non-limiting embodiment, processor 1804 may use a Fast Fourier Transform (FFT) algorithm to compute a discrete Fourier Transform (DFT) of a sequence. Using this frequency domain conversion, processor 1804 may enable the conversion of the initially blurry digitized slide image into a DFT simplified matrix, then applying the FFT to generate a resulting image with improved clarity. Clarity improvements may also be conducted using convolutional neural networks as discussed in reference to other figures herein. 90 Attorney Docket No.1519-029PCT1 Still referring to FIG.18, and in summary, generating and displaying virtual slides using the techniques of the present disclosure may enable various advantages, including but not limited to: (1) ease of visualization of serial sections that have been placed on the same slide; (2) removal or reduction of artifacts such as debris, bubbles, and background stain (or otherwise allowing adjustment of how such artifacts are displayed relative to the remainder of the image); (3) ability to change relative placement of serial sections while creating the virtual slide; (4) reduction of cognitive load when assessing the morphological differences arising because of different orientation between the serial sections in a intra serial section slide; and (5) ease of visualization of intraslide and interslide serial sections derived from same paraffin block stained with different stains for comparative evaluation of morphological features across stains. At a high level, aspects of the present disclosure are directed to apparatus and methods for slide imaging. Apparatus described herein may generate images of slides and/or samples on slides. In an embodiment, said apparatus may capture a first image and identify one or more regions of interest. Regions of interest may include features such as writing, debris, and a sample. A sample may include tissue. In an embodiment, an apparatus may identify a focus pattern of a region of interest. For example, apparatus may identify a focus pattern as a plane based on a plurality of points at which an optimal focus is determined. In an embodiment, apparatus may determine which regions of interest contain a sample. In some embodiments, which regions contain a sample may be determined after one or more other steps described herein. In some embodiments, delaying a determination as to which regions contain a sample may make a slide imaging process more efficient. For example, efficiency gains may result from the difficulty of running sophisticated models on a scanning device during a scan. In another example, efficiency gains may result from minimizing the risk of false positives when classifying a region as debris or annotation and skipping over a scan of a region. In another example, performing steps in this order may be optimal because scanning annotations may be useful. Exemplary embodiments illustrating aspects of the present disclosure are described below in the context of several specific examples. Referring now to FIG.19, an exemplary embodiment of an apparatus 1900 for slide imaging. Apparatus 1900 may include a computing device. Apparatus 1900 may include a processor 1904. Processor 1904 may include, without limitation, any processor 1904 described in this disclosure. Processor 1904 may be included in a computing device. Apparatus 1900 may 91 Attorney Docket No.1519-029PCT1 include at least a processor 1904 and a memory 1908 communicatively connected to the at least a processor 1904, the memory 1908 containing instructions 1912 configuring the at least a processor 1904 to perform one or more processes described herein. Computing device may include any computing device as described in this disclosure, including without limitation a microcontroller, microprocessor, digital signal processor (DSP) and/or system on a chip (SoC) as described in this disclosure. Computing device may include, be included in, and/or communicate with a mobile device such as a mobile telephone or smartphone. Computing device may include a single computing device operating independently, or may include two or more computing device operating in concert, in parallel, sequentially or the like; two or more computing devices may be included together in a single computing device or in two or more computing devices. Computing device may interface or communicate with one or more additional devices as described below in further detail via a network interface device. Network interface device may be utilized for connecting computing device to one or more of a variety of networks, and one or more devices. Examples of a network interface device include, but are not limited to, a network interface card (e.g., a mobile network interface card, a LAN card), a modem, and any combination thereof. Examples of a network include, but are not limited to, a wide area network (e.g., the Internet, an enterprise network), a local area network (e.g., a network associated with an office, a building, a campus or other relatively small geographic space), a telephone network, a data network associated with a telephone/voice provider (e.g., a mobile communications provider data and/or voice network), a direct connection between two computing devices, and any combinations thereof. A network may employ a wired and/or a wireless mode of communication. In general, any network topology may be used. Information (e.g., data, software etc.) may be communicated to and/or from a computer and/or a computing device. Computing device may include but is not limited to, for example, a computing device or cluster of computing devices in a first location and a second computing device or cluster of computing devices in a second location. Computing device may include one or more computing devices dedicated to data storage, security, distribution of traffic for load balancing, and the like. Computing device may distribute one or more computing tasks as described below across a plurality of computing devices of computing device, which may operate in parallel, in series, redundantly, or in any other manner used for distribution of tasks or memory between computing 92 Attorney Docket No.1519-029PCT1 devices. Computing device may be implemented, as a non-limiting example, using a “shared nothing” architecture. With continued reference to FIG.19, computing device may be designed and/or configured to perform any method, method step, or sequence of method steps in any embodiment described in this disclosure, in any order and with any degree of repetition. For instance, computing device may be configured to perform a single step or sequence repeatedly until a desired or commanded outcome is achieved; repetition of a step or a sequence of steps may be performed iteratively and/or recursively using outputs of previous repetitions as inputs to subsequent repetitions, aggregating inputs and/or outputs of repetitions to produce an aggregate result, reduction or decrement of one or more variables such as global variables, and/or division of a larger processing task into a set of iteratively addressed smaller processing tasks. Computing device may perform any step or sequence of steps as described in this disclosure in parallel, such as simultaneously and/or substantially simultaneously performing a step two or more times using two or more parallel threads, processor cores, or the like; division of tasks between parallel threads and/or processes may be performed according to any protocol suitable for division of tasks between iterations. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which steps, sequences of steps, processing tasks, and/or data may be subdivided, shared, or otherwise dealt with using iteration, recursion, and/or parallel processing. Still referring to FIG.19, as used in this disclosure, “communicatively connected” means connected by way of a connection, attachment or linkage between two or more relata which allows for reception and/or transmittance of information therebetween. For example, and without limitation, this connection may be wired or wireless, direct or indirect, and between two or more components, circuits, devices, systems, and the like, which allows for reception and/or transmittance of data and/or signal(s) therebetween. Data and/or signals therebetween may include, without limitation, electrical, electromagnetic, magnetic, video, audio, radio and microwave data and/or signals, combinations thereof, and the like, among others. A communicative connection may be achieved, for example and without limitation, through wired or wireless electronic, digital or analog, communication, either directly or by way of one or more intervening devices or components. Further, communicative connection may include electrically coupling or connecting at least an output of one device, component, or circuit to at least an input 93 Attorney Docket No.1519-029PCT1 of another device, component, or circuit. For example, and without limitation, via a bus or other facility for intercommunication between elements of a computing device. Communicative connecting may also include indirect connections via, for example and without limitation, wireless connection, radio communication, low power wide area network, optical communication, magnetic, capacitive, or optical coupling, and the like. In some instances, the terminology “communicatively coupled” may be used in place of communicatively connected in this disclosure. Still referring to FIG.19, in some embodiments, apparatus 1900 may be used to generate an image of slide 1916 and/or a sample on slide 1916. As used herein, a “slide” is a container or surface holding a sample of interest. In some embodiments, slide 1916 may include a glass slide. In some embodiments, slide 1916 may include a formalin fixed paraffin embedded slide. In some embodiments, a sample on slide 1916 may be stained. In some embodiments, slide 1916 may be substantially transparent. In some embodiments, slide 1916 may include a thin, flat, and substantially transparent glass slide. In some embodiments, a transparent cover may be applied to slide 1916 such that a sample is between slide 1916 and this cover. A sample may include, in non-limiting examples, a blood smear, pap smear, body fluids, and non-biologic samples. In some embodiments, a sample on slide 1916 may include tissue. In some embodiments, sample on slide 1916 may be frozen. Still referring to FIG.19, in some embodiments, slide 1916 and/or a sample on slide 1916 may be illuminated. In some embodiments, apparatus 1900 may include a light source. As used herein, a “light source” is any device configured to emit electromagnetic radiation. In some embodiments, light source may emit a light having substantially one wavelength. In some embodiments, light source may emit a light having a wavelength range. Light source may emit, without limitation, ultraviolet light, visible light, and/or infrared light. In non-limiting examples, light source may include a light-emitting diode (LED), an organic LED (OLED) and/or any other light emitter. Such a light source may be configured to illuminate slide 1916 and/or sample on slide 1916. In a non-limiting example, light source may illuminate slide 1916 and/or sample on slide 1916 from below. Still referring to FIG.19, in some embodiments, apparatus 1900 may include at least an optical system. As used in this disclosure, an "optical system" is an arrangement of one or more components which together act upon or employ electromagnetic radiation. In non-limiting 94 Attorney Docket No.1519-029PCT1 examples, electromagnetic radiation may include light, such as visible light, infrared light, UV light, and the like. An optical system may include one or more optical elements, including without limitation lenses, mirrors, windows, filters, and the like. An optical system may form an optical image that corresponds to an optical object. For instance, an optical system may form an optical image at or upon an optical sensor, which can capture, e.g., digitize, the optical image. In some cases, optical system may have at least a magnification. For instance, optical system may include an objective (e.g., microscope objective) and one or more reimaging optical elements that together produce an optical magnification. In some cases, optical magnification may be referred to herein as zoom. As used herein, an “optical sensor” is a device that measures light and converts the measured light into one or more signals; one or more signals may include, without limitation, one or more electrical signals. In some embodiments, optical sensor 1920 may include at least a photodetector. As used herein, a “photodetector” is a device that is sensitive to light and thereby able to detect light. In some embodiments, a photodetector may include a photodiode, a photoresistor, a photosensor, a photovoltaic chip, and the like. In some embodiments, optical sensor 1920 may include a plurality of photodetectors. Optical sensor 1920 may include, without limitation, a camera. Optical sensor 1920 may be in electronic communication with at least a processor 1904 of apparatus 1900. As used herein, “electronic communication” as used in this disclosure is a shared data connection between two or more devices. In some embodiments, apparatus 1900 may include two or more optical sensors 1920. Still referring to FIG.19, as used herein, “image data” is information representing at least a physical scene, space, and/or object. Image data may include, for example, information representing a sample, slide 1916, or region of a sample or slide. In some cases, image data may be generated by a camera. “Image data” may be used interchangeably through this disclosure with “image,” where image is used as a noun. An image may be optical, such as without limitation where at least an optic is used to generate an image of an object. An image may be digital, such as without limitation when represented as a bitmap. Alternatively, an image may include any media capable of representing a physical scene, space, and/or object. Alternatively, where “image” is used as a verb, in this disclosure, it refers to generation and/or formation of an image. Still referring to FIG.19, in some embodiments, apparatus 1900 may include a slide port 1940. In some embodiments, slide port 1940 may be configured to hold slide 1916. In some 95 Attorney Docket No.1519-029PCT1 embodiments, slide port 1940 may include one or more alignment features. As used herein, an “alignment feature” is a physical feature that helps to secure a slide in place and/or align a slide with another component of an apparatus. In some embodiments, alignment feature may include a component which keeps slide 1916 secure, such as a clamp, latch, clip, recessed area, or another fastener. In some embodiments, slide port 1940 may allow for easy removal or insertion of slide 1916. In some embodiments, slide port 1940 may include a transparent surface through which light may travel. In some embodiments, slide 1916 may rest on and/or may be illuminated by light traveling through such a transparent surface. In some embodiments, slide port 1940 may be mechanically connected to an actuator mechanism 1924 as described below. Still referring to FIG.19, in some embodiments, apparatus 1900 may include an actuator mechanism 1924. As used herein, an “actuator mechanism” is a mechanical component configured to change the relative position of a slide and an optical system. In some embodiments, actuator mechanism 1924 may be mechanically connected to slide 1916, such as slide 1916 in slide port 1940. In some embodiments, actuator mechanism 1924 may be mechanically connected to slide port 1940. For example, actuator mechanism 1924 may move slide port 1940 in order to move slide 1916. In some embodiments, actuator mechanism 1924 may be mechanically connected to at least an optical system. In some embodiments, actuator mechanism 1924 may be mechanically connected to a mobile element. As used herein, a “mobile element” refers to any movable or portable object, component, and device within apparatus 1900 such as, without limitation, a slide, a slide port, or an optical system. In some embodiments, a mobile element may move such that optical system is positioned correctly with respect to slide 1916 such that optical system may capture an image of slide 1916 according to a parameter set. In some embodiments, actuator mechanism 1924 may be mechanically connected to an item selected from the list consisting of slide port 1940, slide 1916, and at least an optical system. In some embodiments, actuator mechanism 1924 may be configured to change the relative position of slide 1916 and optical system by moving slide port 1940, slide 1916, and/or optical system. Still referring to FIG.19, actuator mechanism 1924 may include a component of a machine that is responsible for moving and/or controlling a mechanism or system. Actuator mechanism 1924 may, in some embodiments, require a control signal and/or a source of energy or power. In some cases, a control signal may be relatively low energy. Exemplary control signal forms include electric potential or current, pneumatic pressure or flow, or hydraulic fluid 96 Attorney Docket No.1519-029PCT1 pressure or flow, mechanical force/torque or velocity, or even human power. In some cases, an actuator may have an energy or power source other than control signal. This may include a main energy source, which may include for example electric power, hydraulic power, pneumatic power, mechanical power, and the like. In some embodiments, upon receiving a control signal, actuator mechanism 1924 responds by converting source power into mechanical motion. In some cases, actuator mechanism 1924 may be understood as a form of automation or automatic control. Still referring to FIG.19, in some embodiments, actuator mechanism 1924 may include a hydraulic actuator. A hydraulic actuator may consist of a cylinder or fluid motor that uses hydraulic power to facilitate mechanical operation. Output of hydraulic actuator mechanism 1924 may include mechanical motion, such as without limitation linear, rotatory, or oscillatory motion. In some embodiments, hydraulic actuator may employ a liquid hydraulic fluid. As liquids, in some cases, are incompressible, a hydraulic actuator can exert large forces. Additionally, as force is equal to pressure multiplied by area, hydraulic actuators may act as force transformers with changes in area (e.g., cross sectional area of cylinder and/or piston). An exemplary hydraulic cylinder may consist of a hollow cylindrical tube within which a piston can slide. In some cases, a hydraulic cylinder may be considered single acting. Single acting may be used when fluid pressure is applied substantially to just one side of a piston. Consequently, a single acting piston can move in only one direction. In some cases, a spring may be used to give a single acting piston a return stroke. In some cases, a hydraulic cylinder may be double acting. Double acting may be used when pressure is applied substantially on each side of a piston; any difference in resultant force between the two sides of the piston causes the piston to move. Still referring to FIG.19, in some embodiments, actuator mechanism 1924 may include a pneumatic actuator mechanism 1924. In some cases, a pneumatic actuator may enable considerable forces to be produced from relatively small changes in gas pressure. In some cases, a pneumatic actuator may respond more quickly than other types of actuators, for example hydraulic actuators. A pneumatic actuator may use compressible fluid (e.g., air). In some cases, a pneumatic actuator may operate on compressed air. Operation of hydraulic and/or pneumatic actuators may include control of one or more valves, circuits, fluid pumps, and/or fluid manifolds. 97 Attorney Docket No.1519-029PCT1 Still referring to FIG.19, in some cases, actuator mechanism 1924 may include an electric actuator. Electric actuator mechanism 1924 may include any of electromechanical actuators, linear motors, and the like. In some cases, actuator mechanism 1924 may include an electromechanical actuator. An electromechanical actuator may convert a rotational force of an electric rotary motor into a linear movement to generate a linear movement through a mechanism. Exemplary mechanisms, include rotational to translational motion transformers, such as without limitation a belt, a screw, a crank, a cam, a linkage, a scotch yoke, and the like. In some cases, control of an electromechanical actuator may include control of electric motor, for instance a control signal may control one or more electric motor parameters to control electromechanical actuator. Exemplary non-limitation electric motor parameters include rotational position, input torque, velocity, current, and potential. Electric actuator mechanism 1924 may include a linear motor. Linear motors may differ from electromechanical actuators, as power from linear motors is output directly as translational motion, rather than output as rotational motion and converted to translational motion. In some cases, a linear motor may cause lower friction losses than other devices. Linear motors may be further specified into at least 3 different categories, including flat linear motor, U-channel linear motors and tubular linear motors. Linear motors may be directly controlled by a control signal for controlling one or more linear motor parameters. Exemplary linear motor parameters include without limitation position, force, velocity, potential, and current. Still referring to FIG.19, in some embodiments, an actuator mechanism 1924 may include a mechanical actuator mechanism 1924. In some cases, a mechanical actuator mechanism 1924 may function to execute movement by converting one kind of motion, such as rotary motion, into another kind, such as linear motion. An exemplary mechanical actuator includes a rack and pinion. In some cases, a mechanical power source, such as a power take off may serve as power source for a mechanical actuator. Mechanical actuators may employ any number of mechanisms, including for example without limitation gears, rails, pulleys, cables, linkages, and the like. Still referring to FIG.19, in some embodiments, actuator mechanism 1924 may be in electronic communication with actuator controls. As used herein, “actuator controls” is a system configured to operate actuator mechanism such that a slide and an optical system reach a desired relative position. In some embodiments, actuator controls may operate actuator mechanism 1924 98 Attorney Docket No.1519-029PCT1 based on input received from a user interface 1936. In some embodiments, actuator controls may be configured to operate actuator mechanism 1924 such that optical system is in a position to capture an image of an entire sample. In some embodiments, actuator controls may be configured to operate actuator mechanism 1924 such that optical system is in a position to capture an image of a region of interest, a particular horizontal row, a particular point, a particular focus depth, and the like. Electronic communication between actuator mechanism 1924 and actuator controls may include transmission of signals. For example, actuator controls may generate physical movements of actuator mechanism in response to an input signal. In some embodiments, input signal may be received by actuator controls from processor 1904 or input interface 1928. Still referring to FIG.19, as used in this disclosure, a “signal” is any intelligible representation of data, for example from one device to another. A signal may include an optical signal, a hydraulic signal, a pneumatic signal, a mechanical signal, an electric signal, a digital signal, an analog signal, and the like. In some cases, a signal may be used to communicate with a computing device, for example by way of one or more ports. In some cases, a signal may be transmitted and/or received by a computing device, for example by way of an input/output port. An analog signal may be digitized, for example by way of an analog to digital converter. In some cases, an analog signal may be processed, for example by way of any analog signal processing steps described in this disclosure, prior to digitization. In some cases, a digital signal may be used to communicate between two or more devices, including without limitation computing devices. In some cases, a digital signal may be communicated by way of one or more communication protocols, including without limitation internet protocol (IP), controller area network (CAN) protocols, serial communication protocols (e.g., universal asynchronous receiver-transmitter [UART]), parallel communication protocols (e.g., IEEE 128 [printer port]), and the like. Still referring to FIG.19, in some embodiments, apparatus 1900 may perform one or more signal processing steps on a signal. For instance, apparatus 1900 may analyze, modify, and/or synthesize a signal representative of data in order to improve the signal, for instance by improving transmission, storage efficiency, or signal to noise ratio. Exemplary methods of signal processing may include analog, continuous time, discrete, digital, nonlinear, and statistical. Analog signal processing may be performed on non-digitized or analog signals. Exemplary analog processes may include passive filters, active filters, additive mixers, integrators, delay 99 Attorney Docket No.1519-029PCT1 lines, compandors, multipliers, voltage-controlled filters, voltage-controlled oscillators, and phase-locked loops. Continuous-time signal processing may be used, in some cases, to process signals which vary continuously within a domain, for instance time. Exemplary non-limiting continuous time processes may include time domain processing, frequency domain processing (Fourier transform), and complex frequency domain processing. Discrete time signal processing may be used when a signal is sampled non-continuously or at discrete time intervals (i.e., quantized in time). Analog discrete-time signal processing may process a signal using the following exemplary circuits sample and hold circuits, analog time-division multiplexers, analog delay lines and analog feedback shift registers. Digital signal processing may be used to process digitized discrete-time sampled signals. Commonly, digital signal processing may be performed by a computing device or other specialized digital circuits, such as without limitation an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a specialized digital signal processor (DSP). Digital signal processing may be used to perform any combination of typical arithmetical operations, including fixed-point and floating-point, real- valued and complex-valued, multiplication and addition. Digital signal processing may additionally operate circular buffers and lookup tables. Further non-limiting examples of algorithms that may be performed according to digital signal processing techniques include fast Fourier transform (FFT), finite impulse response (FIR) filter, infinite impulse response (IIR) filter, and adaptive filters such as the Wiener and Kalman filters. Statistical signal processing may be used to process a signal as a random function (i.e., a stochastic process), utilizing statistical properties. For instance, in some embodiments, a signal may be modeled with a probability distribution indicating noise, which then may be used to reduce noise in a processed signal. Still referring to FIG.19, in some embodiments, apparatus 1900 may include a user interface 1936. User interface 1936 may include output interface 1932 and input interface 1928. Still referring to FIG.19, in some embodiments, output interface 1932 may include one or more elements through which apparatus 1900 may communicate information to a user. In a non-limiting example, output interface 1932 may include a display. A display may include a high resolution display. A display may output images, videos, and the like to a user. In another non- limiting example, output interface 1932 may include a speaker. A speaker may output audio to a 100 Attorney Docket No.1519-029PCT1 user. In another non-limiting example, output interface 1932 may include a haptic device. A speaker may output haptic feedback to a user. Still referring to FIG.19, in some embodiments, optical system may include a camera. In some cases, a camera may include one or more optics. Exemplary non-limiting optics include spherical lenses, aspherical lenses, reflectors, polarizers, filters, windows, aperture stops, and the like. In some embodiments, one or more optics associated with a camera may be adjusted in order to, in non-limiting examples, change the zoom, depth of field, and/or focus distance of the camera. In some embodiments, one or more of such settings may be configured to detect a feature of a sample on slide 1916. In some embodiments, one or more of such settings may be configured based on a parameter set, as described below. In some embodiments, camera may capture images at a low depth of field. In a non-limiting example, camera may capture images such that a first depth of sample is in focus and a second depth of sample is out of focus. In some embodiments, an autofocus mechanism may be used to determine focus distance. In some embodiments, focus distance may be set by parameter set. In some embodiments, camera may be configured to capture a plurality of images at different focus distances. In a non-limiting example, camera may capture a plurality of images at different focus distances, such that images are captured where each focus depth of the sample is in focus in at least one image. In some embodiments, at least a camera may include an image sensor. Exemplary non-limiting image sensors include digital image sensors, such as without limitation charge-coupled device (CCD) sensors and complimentary metal-oxide-semiconductor (CMOS) sensors. In some embodiments, a camera may be sensitive within a non-visible range of electromagnetic radiation, such as without limitation infrared. Still referring to FIG.19, in some embodiments, input interface 1928 may include controls for operating apparatus 1900. Such controls may be operated by a user. Input interface 1928 may include, in non-limiting examples, a camera, microphone, keyboard, touch screen, mouse, joystick, foot pedal, button, dial, and the like. Input interface 1928 may accept, in non- limiting examples, mechanical input, audio input, visual input, text input, and the like. In some embodiments, audio inputs into input interface 1928 may be interpreted using an automatic speech recognition function, allowing a user to control apparatus 1900 via speech. In some embodiments, input interface 1928 may approximate controls of a microscope. 101 Attorney Docket No.1519-029PCT1 Still referring to FIG.19, in some embodiments, audio inputs may be processed using automatic speech recognition. In some embodiments, automatic speech recognition may require training (i.e., enrollment). In some cases, training an automatic speech recognition model may require an individual speaker to read text or isolated vocabulary. In some cases, audio training data may include an audio component having an audible verbal content, the contents of which are known a priori by a computing device. Computing device may then train an automatic speech recognition model according to training data which includes audible verbal content correlated to known content. In this way, computing device may analyze a person's specific voice and train an automatic speech recognition model to the person's speech, resulting in increased accuracy. Alternatively, or additionally, in some cases, computing device may include an automatic speech recognition model that is speaker independent. As used in this disclosure, a “speaker independent” automatic speech recognition process does not require training for each individual speaker. Conversely, as used in this disclosure, automatic speech recognition processes that employ individual speaker specific training are “speaker dependent.” Still referring to FIG.19, in some embodiments, an automatic speech recognition process may perform voice recognition or speaker identification. As used in this disclosure, “voice recognition” refers to identifying a speaker, from audio content, rather than what the speaker is saying. In some cases, computing device may first recognize a speaker of verbal audio content and then automatically recognize speech of the speaker, for example by way of a speaker dependent automatic speech recognition model or process. In some embodiments, an automatic speech recognition process can be used to authenticate or verify an identity of a speaker. In some cases, a speaker may or may not include subject. For example, subject may speak within audio inputs, but others may speak as well. Still referring to FIG.19, in some embodiments, an automatic speech recognition process may include one or all of acoustic modeling, language modeling, and statistically based speech recognition algorithms. In some cases, an automatic speech recognition process may employ hidden Markov models (HMMs). As discussed in greater detail below, language modeling such as that employed in natural language processing applications like document classification or statistical machine translation, may also be employed by an automatic speech recognition process. 102 Attorney Docket No.1519-029PCT1 Still referring to FIG.19, an exemplary algorithm employed in automatic speech recognition may include or even be based upon hidden Markov models. Hidden Markov models (HMMs) may include statistical models that output a sequence of symbols or quantities. HMMs can be used in speech recognition because a speech signal can be viewed as a piecewise stationary signal or a short-time stationary signal. For example, over a short time scale (e.g., 10 milliseconds), speech can be approximated as a stationary process. Speech (i.e., audible verbal content) can be understood as a Markov model for many stochastic purposes. Still referring to FIG.19, in some embodiments HMMs can be trained automatically and may be relatively simple and computationally feasible to use. In an exemplary automatic speech recognition process, a hidden Markov model may output a sequence of n-dimensional real- valued vectors (with n being a small integer, such as 10), at a rate of about one vector every 10 milliseconds. Vectors may consist of cepstral coefficients. A cepstral coefficient requires using a spectral domain. Cepstral coefficients may be obtained by taking a Fourier transform of a short time window of speech yielding a spectrum, decorrelating the spectrum using a cosine transform, and taking first (i.e., most significant) coefficients. In some cases, an HMM may have in each state a statistical distribution that is a mixture of diagonal covariance Gaussians, yielding a likelihood for each observed vector. In some cases, each word, or phoneme, may have a different output distribution; an HMM for a sequence of words or phonemes may be made by concatenating an HMMs for separate words and phonemes. Still referring to FIG.19, in some embodiments, an automatic speech recognition process may use various combinations of a number of techniques in order to improve results. In some cases, a large-vocabulary automatic speech recognition process may include context dependency for phonemes. For example, in some cases, phonemes with different left and right context may have different realizations as HMM states. In some cases, an automatic speech recognition process may use cepstral normalization to normalize for different speakers and recording conditions. In some cases, an automatic speech recognition process may use vocal tract length normalization (VTLN) for male-female normalization and maximum likelihood linear regression (MLLR) for more general speaker adaptation. In some cases, an automatic speech recognition process may determine so-called delta and delta-delta coefficients to capture speech dynamics and might use heteroscedastic linear discriminant analysis (HLDA). In some cases, an automatic speech recognition process may use splicing and a linear discriminate analysis (LDA)-based 103 Attorney Docket No.1519-029PCT1 projection, which may include heteroscedastic linear discriminant analysis or a global semi-tied covariance transform (also known as maximum likelihood linear transform [MLLT]). In some cases, an automatic speech recognition process may use discriminative training techniques, which may dispense with a purely statistical approach to HMM parameter estimation and instead optimize some classification-related measure of training data; examples may include maximum mutual information (MMI), minimum classification error (MCE), and minimum phone error (MPE). Still referring to FIG.19, in some embodiments, an automatic speech recognition process may be said to decode speech (i.e., audible verbal content). Decoding of speech may occur when an automatic speech recognition system is presented with a new utterance and must compute a most likely sentence. In some cases, speech decoding may include a Viterbi algorithm. A Viterbi algorithm may include a dynamic programming algorithm for obtaining a maximum a posteriori probability estimate of a most likely sequence of hidden states (i.e., Viterbi path) that results in a sequence of observed events. Viterbi algorithms may be employed in context of Markov information sources and hidden Markov models. A Viterbi algorithm may be used to find a best path, for example using a dynamically created combination hidden Markov model, having both acoustic and language model information, using a statically created combination hidden Markov model (e.g., finite state transducer [FST] approach). Still referring to FIG.19, in some embodiments, speech (i.e., audible verbal content) decoding may include considering a set of good candidates and not only a best candidate, when presented with a new utterance. In some cases, a better scoring function (i.e., re-scoring) may be used to rate each of a set of good candidates, allowing selection of a best candidate according to this refined score. In some cases, a set of candidates can be kept either as a list (i.e., N-best list approach) or as a subset of models (i.e., a lattice). In some cases, re-scoring may be performed by optimizing Bayes risk (or an approximation thereof). In some cases, re-scoring may include optimizing for sentence (including keywords) that minimizes an expectancy of a given loss function with regards to all possible transcriptions. For example, re-scoring may allow selection of a sentence that minimizes an average distance to other possible sentences weighted by their estimated probability. In some cases, an employed loss function may include Levenshtein distance, although different distance calculations may be performed, for instance for specific tasks. In some cases, a set of candidates may be pruned to maintain tractability. 104 Attorney Docket No.1519-029PCT1 Still referring to FIG.19, in some embodiments, an automatic speech recognition process may employ dynamic time warping (DTW)-based approaches. Dynamic time warping may include algorithms for measuring similarity between two sequences, which may vary in time or speed. For instance, similarities in walking patterns would be detected, even if in one video the person was walking slowly and if in another he or she were walking more quickly, or even if there were accelerations and deceleration during the course of one observation. DTW has been applied to video, audio, and graphics – indeed, any data that can be turned into a linear representation can be analyzed with DTW. In some cases, DTW may be used by an automatic speech recognition process to cope with different speaking (i.e., audible verbal content) speeds. In some cases, DTW may allow computing device to find an optimal match between two given sequences (e.g., time series) with certain restrictions. That is, in some cases, sequences can be "warped" non-linearly to match each other. In some cases, a DTW-based sequence alignment method may be used in context of hidden Markov models. Still referring to FIG.19, in some embodiments, an automatic speech recognition process may include a neural network. Neural network may include any neural network, for example those disclosed with reference to other figures. In some cases, neural networks may be used for automatic speech recognition, including phoneme classification, phoneme classification through multi-objective evolutionary algorithms, isolated word recognition, audiovisual speech recognition, audiovisual speaker recognition and speaker adaptation. In some cases, neural networks employed in automatic speech recognition may make fewer explicit assumptions about feature statistical properties than HMMs and therefore may have several qualities making them attractive recognition models for speech recognition. When used to estimate the probabilities of a speech feature segment, neural networks may allow discriminative training in a natural and efficient manner. In some cases, neural networks may be used to effectively classify audible verbal content over short-time interval, for instance such as individual phonemes and isolated words. In some embodiments, a neural network may be employed by automatic speech recognition processes for pre-processing, feature transformation and/or dimensionality reduction, for example prior to HMM-based recognition. In some embodiments, long short-term memory (LSTM) and related recurrent neural networks (RNNs) and Time Delay Neural Networks (TDNN's) may be used for automatic speech recognition, for example over longer time intervals for continuous speech recognition. 105 Attorney Docket No.1519-029PCT1 Still referring to FIG.19, in some embodiments, apparatus 1900 captures a first image of slide 1916 at a first position. In some embodiments, first image may be captured using at least an optical system. Still referring to FIG.19, in some embodiments, capturing a first image of a slide 1916 at a first position may include using actuator mechanism 1924 and/or actuator controls to move optical system and/or slide 1916 into desired positions. In some embodiments, first image includes an image of the entire sample and/or the entire slide 1916. In some embodiments, first image includes an image of a region of a sample. In some embodiments, first image includes a wider angle image than second image (described below). In some embodiments, first image may include a lower resolution image than second image. Still referring to FIG.19, in some embodiments, apparatus 1900 may identify at least a region of interest within first image. In some embodiments, machine vision may be used to identify at least a region of interest. As used herein, a “region of interest” is a specific area within a slide, or a digital image of a slide, in which a feature is detected. A feature may include, in non-limiting examples, a sample, debris, writing on a slide, a crack in a slide, a bubble, and the like. Still referring to FIG.19, in some embodiments, apparatus 1900 may include machine learning module 1944. In some embodiments, apparatus 1900 may use ROI identification machine learning model 1948 to identify at least a region of interest. In some embodiments, ROI identification machine learning model 1948 may be trained using supervised learning. In some embodiments, ROI identification machine learning model 1948 may include a classifier. ROI identification machine learning model 1948 may be trained on a dataset including example images of slides, associated with example regions of those images in which a feature is present. Such a training dataset may be gathered by, for example, gathering data slide imaging devices as to which regions of images of slides professionals focus or zoom in on. Once trained, ROI identification machine learning model 1948 may accept as an input an image of a slide and may output data as to the location of any regions of interest present. In some embodiments, a neural network, such as a convolutional neural network, may be used to identify at least a region of interest. For example, a convolutional neural network may be used to detect edges in an image of a slide, and at least a region of interest may be identified based on the presence of edges. In some embodiments, at least a region of interest may be identified as a function of differences in 106 Attorney Docket No.1519-029PCT1 brightness and/or color in comparison to a background brightness and/or color. In some embodiments, apparatus 1900 may identify a region of interest using a classifier. In some embodiments, segments of an image may be input into a classifier, and the classifier may categorize them based on whether a region of interest is present. In some embodiments, a classifier may output a score indicating the degree to which a region of interest is detected and/or a confidence level that a region of interest is present. In some embodiments, a feature may be detected using a neural network or other machine learning model trained to detect a feature and/or object. For example, edges, corners, blobs, or ridges may be detected, and whether a location is determined to be within a region of interest may depend on detection of such elements. In some embodiments, a machine learning model such as a support vector machine technique may be used to determine a feature based on detection of elements such as edges, corners, blobs, or ridges. Still referring to FIG.19, in some embodiments, apparatus 1900 may identify at least a region of interest as a function of user input. For example, a user may modify a setting on a degree of sensitivity with which to detect regions of interest. In this example, if a user input indicates a low degree of sensitivity, then apparatus 1900 may only detect major regions of interest. This may involve, for example, ignoring potential regions of interest that are below a certain size. In another example, this may involve applying a machine learning model such as a classifier to an image (or segment of an image) and identifying it as a region of interest only if the machine learning model outputs a score higher than a threshold, where the score indicates the degree to which a region of interest is detected and/or a confidence level that a region of interest is present. In some embodiments, an image may be split into smaller segments and segments may be analyzed in order to determine whether at least a region of interest is present. Still referring to FIG.19, in some embodiments, apparatus 1900 may receive at least a region of interest. In some embodiments, apparatus 1900 may receive at least a region of interest without first capturing a first image. In a non-limiting example, a user may input a region of interest. In some embodiments, apparatus 1900 may capture a first image, and may apply a received region of interest to the first image. This may be done, for example, where a region of interest is received before first image is captured. In some embodiments, apparatus 1900 may capture a first image as a function of a region of interest. In a non-limiting example, apparatus 107 Attorney Docket No.1519-029PCT1 1900 may receive a region of interest from a user through user input, and may capture a first image at a first position within the region of interest. Still referring to FIG.19, apparatus 1900 may identify a focus pattern. In some embodiments, apparatus 1900 may identify a focus pattern in at least a region of interest, such as in each region of interest detected as described herein. In some embodiments, identifying a focus pattern may include identifying a row, identifying a point within a row, determining an optimal focus at a point, and/or identifying a plane. Still referring to FIG.19, as used herein, a “row” of a digital image of a slide is a segment of a digital image of a slide between two parallel lines. A row may include, for example, a line of pixels with a width of 1 pixel. In another example, a row may have a width of multiple pixels. A row may or may not travel diagonally across a grid of pixels. As used herein, a “point” on a digital image of a slide is a specific position within a digital image of a slide. For example, in a digital image made up of a grid of pixels, a point may have a specific (x,y) position. As used herein, an “optimal focus” is a focus distance at which the subject that is being focused on is in focus. Still referring to FIG.19, in some embodiments, apparatus 1900 may identify a row within a region of interest. A row may be identified based on a first row sample presence score. A first row sample presence score may be identified using machine vision. A first row sample presence score may be identified based on an output of row identification machine learning model 1952. A first row sample presence score may be identified by determining one or more row sample presence scores for adjacent rows. For example, a first row sample presence score may be determined as a function of a second row sample presence score and a third row sample presence score, where the second and third row sample presence scores are based on rows adjacent to the row of the first row sample presence score. As used herein, a “sample presence score” is a value representing or estimating the likelihood that sample is present in a location. As used herein, a “row sample presence score” is a sample presence score where the location is a row. A sample presence score need not represent sample presence likelihood as a percentage on a 0-100% scale. For example, a first row may have a row sample presence score of 2000 and a second row may have a row sample presence score of 3000, indicating that the second row is more likely to contain sample than the first row. A first row sample presence score may be identified, in non-limiting examples, based on a sum of row sample presence scores for adjacent 108 Attorney Docket No.1519-029PCT1 rows and/or a weighted sum of row sample presence scores for adjacent rows and/or a minimum row sample presence score for adjacent rows. In some embodiments, a first set of row sample presence scores is identified using a machine learning model, such as row identification machine learning model 1952, and a second set of row sample presence scores is identified based on row sample presence scores for adjacent rows from the first set of sample presence scores. In some embodiments, such a second set of row sample presence scores may be used to identify a best row. A row sample presence score may be identified by determining a sample presence score for rows not immediately adjacent to a row in question. For example, row sample presence scores may be determined for a row in question, adjacent rows, and rows separated from the row in question by a single row, and each of these row sample presence scores may factor into identifying a row (for example, using a weighted sum of sample presence scores). A sample presence score may be determined using machine vision. In some embodiments, a row sample presence score may be determined based on a section of a row that does not fully cross an image and/or slide. For example, a row sample presence score may be determined for the width of a region of interest. In another example, a row sample presence score may be determined for a section of a row with a certain pixel width. Still referring to FIG.19, in some embodiments, identifying a focus pattern may include identifying a row which includes a specific (X,Y) position, such as a position with which an optimal focus distance is to be calculated. In some embodiments, identifying a focus pattern may include capturing a plurality of images at such a location, wherein each of the plurality of images has a different focus distance. Such images may represent a z-stack as described further below. Identifying a focus pattern may further include determining an optimally focused image of the plurality of images. A focus distance of such image may be determined to be an optimal focus distance for that (X,Y) position. Optimal focus distances may be determined in this way for a plurality of points in that row. A focus pattern may be identified using focus distances of a plurality of optimally focused images at such plurality of points in the row. Still referring to FIG.19, in some embodiments, a row may further include a second position. Apparatus 1900 may capture a plurality of second images at such second location, wherein each of the plurality of second images have a different focus distance. Apparatus 1900 may determine an optimally focused second image of the plurality of second images having an optimal focus. Apparatus 1900 may identify a focus pattern using the focus distance of an 109 Attorney Docket No.1519-029PCT1 optimally focused first image and the optimally focused second image. For example, a focus pattern may be determined to include a line connecting the two points. Apparatus 1900 may extrapolate a third focus distance for a third position as a function of a focus pattern. In some embodiments, extrapolating third focus distance may include using first or second focus distance as third focus distance. In some embodiments, extrapolation may include linear extrapolation, polynomial extrapolation, conic extrapolation, geometric extrapolation, and the like. In some embodiments, such third position may be located outside of a row including a first position and/or a second position. In some embodiments, such third position may be located within a different region of interest than a first position and/or a second position. Still referring to FIG.19, in some embodiments, a sample presence score may be determined using row identification machine learning model 1952. In some embodiments, row identification machine learning model 1952 may be trained using supervised learning. Row identification machine learning model may be trained on a dataset including example rows from images of slides, associated with whether a sample is present. Such a dataset may be gathered by, for example, capturing images of slides, manually identifying which rows contain a sample, and extracting rows from the larger images. Once row identification machine learning model 1952 is trained, it may accept as an input a row from an image of a slide, such as a row from a region of interest, and may output a determination as to whether a sample is present and/or a sample presence score, such as a row sample presence score. Still referring to FIG.19, in some embodiments, apparatus 1900 may identify a point within a row, such as a row identified as described above. In some embodiments, a point may be identified using machine vision. In some embodiments, a point may be identified based on a point within a particular row that has a maximum point sample presence score. As used herein, a “point sample presence score” is a sample presence score where the location is a point. A point sample presence score may be determined, in non-limiting examples, based on a color of a point and/or surrounding pixels, or whether a point is inside or outside potential specimen boundaries (which may be determined, for example, identifying edges within an image and a region of the image enclosed by edges). In another non-limiting example, a point sample presence score may be determined based on the distance between a point and the edge of a slide. Still referring to FIG.19, in some embodiments, a point may be identified using point identification machine learning model 1956. In some embodiments, point identification machine 110 Attorney Docket No.1519-029PCT1 learning model 1956 may be trained using supervised learning. Point identification machine learning model may be trained on a dataset including example points from images of slides, associated with whether a sample is present. Such a dataset may be gathered by, for example, capturing images of slides, manually identifying which points contain a sample, and extracting points from the larger images. Once point identification machine learning model 1956 is trained, it may accept as an input a point from an image of a slide, such as a point from a region of interest, and may output a determination as to whether a sample is present and/or a point sample presence score. In some embodiments, apparatus 1900 may identify a point from a row based on which one has the highest point sample presence score. Still referring to FIG.19, in some embodiments, apparatus 1900 may determine an optimal focus at a point, such as a point identified as described above. In some embodiments, optimal focus may be determined using an autofocus mechanism. In some embodiments, optimal focus may be determined using a rangefinder. In some embodiments, actuator mechanism 1924 may move optical sensor 1920 and/or slide port 1940 such that an autofocus mechanism focuses on a desired location. In some embodiments, an autofocus mechanism may be capable of focusing on multiple points within a frame and may select which point to focus on based on a point identified as described above. In some embodiments, one or more camera parameters other than focus may be adjusted to improve focus and/or improve an image. For example, aperture may be adjusted to modify the degree to which a point is in focus. For example, aperture may be adjusted to increase a depth of field such that a point is in focus. An optimal focus may be expressed, in non-limiting examples, as a focus distance or a focus depth. Still referring to FIG.19, in some embodiments, apparatus 1900 may identify a focus pattern based on an optimal focus and/or a point. An optimal focus and a point may be expressed as a location in 3 dimensional space. For example, X and Y coordinates (horizontal axes) may be determined based on the location of the point on the slide and/or the location of the point on an image. A Z coordinate may be determined based on an optimal focus. For example, a focus distance of an optimal focus may be used as a Z coordinate. As used herein, a “focus pattern” is a pattern that approximates an optimal level of focus at multiple points, including at least one point at which an optimal focus has not been measured. One or more (X,Y,Z) coordinate sets may be used to determine a focus pattern. One or more default parameters may be used to determine a focus pattern (such as defaulting to horizontal where there is insufficient 111 Attorney Docket No.1519-029PCT1 data to determine otherwise). For example, a single (X,Y,Z) coordinate set may be determined, and a focus pattern may be determined as a plane extending horizontally in the X and Y directions, with a Z level remaining constant. In some embodiments, a focus pattern varies in a vertical (Z) direction. For example, two (X,Y,Z) coordinate sets may be determined, and a focus pattern may be determined as including a line between those two (X,Y,Z) locations, and extending horizontally when moving horizontally perpendicular to the line (to form a plane that includes the line). In another example, three (X,Y,Z) coordinate sets may be determined, and a focus pattern may be determined as a plane including all 3 locations. In another example, several (X,Y,Z) coordinate sets may be determined, and a regression algorithm may be used to determine a plane that best fits the coordinate sets. For example, least squares regression may be used. In some embodiments, a focus pattern is not a plane. A focus pattern may include a surface, such as a surface in 3D space. A focus pattern may include one or more curves, bumps, edges, and the like. For example, a focus pattern may include a first plane in a first region, a second plane in a second region, and an edge where the planes intersect. In another example, a focus pattern may include a curved and/or bumpy surface determined such that the Z level of each location on a focus pattern surface is based on the distance to nearby (X,Y,Z) coordinate sets and their Z levels. In another example, a focus pattern may include a plurality of shapes, with (X,Y,Z) coordinate sets as their vertices and lines between (X,Y,Z) coordinate sets as their borders. In some embodiments, a focus pattern may have only one Z value for each (X,Y) coordinate set. In some embodiments, which points have their optimal focus assessed may be determined as a function of, in non-limiting examples, a desired density of points within a region of interest, a likelihood that a sample is present (such as an output from a ML model described above), and/or a user input. For example, a user may manually select a point and/or may input a desired point density. In some embodiments, a focus pattern may be updated as further points are scanned. For example, a focus pattern may be in the shape of a plane based on 10 (X,Y,Z) coordinate sets, and an 11^th (X,Y,Z) coordinate set may be scanned, and the focus pattern may be recalculated and/or updated to take into account the new coordinate set. In another example, a focus pattern may start as a plane based on a single (X,Y,Z) coordinate set, and may be updated as additional (X,Y,Z) coordinate sets are identified. In some embodiments, a focus pattern may be updated with each additional (X,Y,Z) coordinate set. In some embodiments, a focus pattern may be updated at a rate less than the rate of (X,Y,Z) coordinate set identification. In non- 112 Attorney Docket No.1519-029PCT1 limiting examples, a focus pattern may be updated every 2, 3, 4, 5, or more (X,Y,Z) coordinate sets. Still referring to FIG.19, in some embodiments, data used to identify a focus pattern may be filtered. In some embodiments, one or more outliers may be removed. For example, if nearly all (X,Y,Z) points suggest a focus pattern in the shape of a plane, and a single (X,Y,Z) point has a very different Z value than would be estimated from the plane, then that (X,Y,Z) point may be removed. In another example, (X,Y,Z) points identified as focusing on features other than a sample may be removed. For example, (X,Y,Z) points identified as focusing on annotations may be removed. Still referring to FIG.19, in some embodiments, a process described above may be used to determine a focus pattern in each region of interest. In regions of interest that contain a sample, this may result in determining a focus pattern based on one or more points that include a sample. This may aid in efficiently identifying a focus pattern such that follow up images may be captured at correct focus distances. In some regions of interest, such as regions of interest that do not contain a sample but instead include features such as annotations, this may result in focusing on a non-sample feature. This may be desirable, for example, because capturing in focus images of features such as annotations may aid professionals in reading annotations and/or may aid optical character recognition processes in transcribing writing. In some regions of interest, both sample and non-sample features may be present; in this case, focusing on a sample may be desirable, and processes described herein may achieve this. In some embodiments, a process described herein may offer a more efficient method of identifying a focus pattern than alternatives. For example, a process described herein may require focusing on fewer points in order to determine a focus pattern. Still referring to FIG.19, in some embodiments, a focus pattern, such as a plane, may be identified as a function of a first image and a first position. An (X,Y) position of a point may be determined from the first position. An optimal focus value may be determined from the first image. Together, these may be used to identify an (X,Y,Z) coordinate set which may be used to identify a focus pattern as described herein. Still referring to FIG.19, in some embodiments, a focus pattern, such as a Z level of an optimal focus may be used to scan the rest of a row. For example, this may be used to scan the rest of a row including a point at which the optimal focus was identified. In some embodiments, 113 Attorney Docket No.1519-029PCT1 a focus pattern may be used to scan further rows. For example, a focus pattern determined as a function of a Z level of a point may be used to scan rows adjacent to a row including the point. In some embodiments, a focus pattern may be used to scan other rows within the same region of interest. In some embodiments, a focus pattern may be used to scan rows within other regions of interest, such as regions of interest that are nearby. Still referring to FIG.19, in some embodiments, region of interest identification, row identification, point identification, and/or focus pattern determination may be done locally. For example, apparatus 1900 may include an already trained machine learning model and may apply the model to an image. In some embodiments, region of interest identification, row identification, point identification, and/or focus pattern determination may be done externally. For example, apparatus 1900 may transmit image data to another computing device and may receive an output described herein. In some embodiments, a region of interest may be identified, a row may be identified, a point may be identified, and/or a focus pattern may be determined in real time. Still referring to FIG.19, in some embodiments, apparatus 1900 may determine a scanning pattern. A scanning pattern may be based on, for example, sample morphology. Scanning patterns may include, in non-limiting examples, zig-zag, snake line, and spiral. Still referring to FIG.19, in some embodiments, a snake pattern may be used to scan a slide. A snake pattern may progress in any horizontal direction. In non-limiting examples, a snake pattern may progress across the length, or the width, of a slide. In some embodiments, a direction a snake pattern progresses in may be chosen in order to minimize a number of necessary turns. For example, a snake pattern may progress along the shortest dimension of a slide. In another example, a sample shape may be identified, for example, using machine vision, and a snake pattern may progress in a direction according to the shortest dimension of the sample. In some embodiments, a snake pattern may be chosen when high scan speed is desired. In some embodiments, a snake pattern may minimize movement of a camera and/or a slide needed to scan the slide. In some embodiments, a zig-zag pattern may be used to scan a slide. As described with respect to a snake pattern scan, a zig-zag pattern scan may be performed in any horizontal direction, and a direction may be selected in order to minimize a number of turns and/or rows used to scan a slide and/or a feature such as a sample. In some embodiments, a spiral pattern may be used to scan a slide, feature, region of interest, or the like. In some 114 Attorney Docket No.1519-029PCT1 embodiments, a snake pattern and/or a zig-zag pattern may be optimal for Z translation from one row to the next. In some embodiments, a spiral pattern may be optimal for dynamic grid region of interest extension. Still referring to FIG.19, in some embodiments, apparatus 1900 may extrapolate a focus distance for a second position as a function of a focus pattern. In some embodiments, a focus pattern may be determined as a function of one or more (X,Y,Z) points local to a region of interest and/or a subregion of a region of interest. In some embodiments, a focus pattern such as a plane may be used to approximate an optimal level of focus using extrapolation (such as, rather than interpolation). For example, a focus pattern may be used to approximate an optimal level of focus at an (X,Y) point outside of the range of (X,Y) points already scanned, such as outside of the range of X-values, outside of the range of Y-values, or outside of a shape encompassing the (X,Y) points already scanned. In another example, a focus pattern may be used to approximate an optimal level of focus, where the optimal level of focus includes a Z value outside of the range of Z values used to determine the focus pattern. In another example, an (X,Y,Z) point may be extrapolated to a focus pattern over a row. In some embodiments, one or more additional (X,Y,Z) points may be used to update a focus pattern. In some embodiments, a focus pattern identified for a row may be extrapolated to another row, such as an adjacent row. In another example, a local focus pattern such as a plane may be extrapolated to identify an optimal level of focus outside of a local region. In another example, a focus pattern in a first region of interest may be extrapolated to generate a focus pattern in a second region of interest and/or identify an optimal level of focus at an (X,Y) point in a second region of interest. Still referring to FIG.19, in some embodiments, apparatus 1900 may capture a second image of a slide at a second position, and at a focus distance based on the focus pattern. For example, if a focus pattern is a plane, and an in-focus image of a particular (X,Y) point is desired, then apparatus 1900 may capture an image using a focus distance based on the Z coordinate of the plane at those (X,Y) coordinates. In some embodiments, a first position (such as one at which an optimal focus is measured) and a second position may be set to image locations in the same region of interest. In some embodiments, an actuator mechanism may be mechanically connected to a mobile element; and the actuator mechanism may move the mobile element into the second position. In some embodiments, obtaining a second image may include capturing a plurality of images taken with focus distances based on a focus pattern and 115 Attorney Docket No.1519-029PCT1 constructing the second image from the plurality of images. In some embodiments, a second image may include an image taken as part of a z-stack. Z-stacks are described further below. Still referring to FIG.19, in some embodiments, apparatus 1900 may determine which regions of interest contain a sample. In some embodiments, this may be applied to a second image, such as a second image taken using a focus distance based on a focus pattern. In some embodiments, which regions of interest may be determined using sample identification machine learning model 1960. In some embodiments, sample identification machine learning model 1960 may include a classifier. In some embodiments, sample identification machine learning model 1960 may be trained using supervised learning. Sample identification machine learning model 1960 may be trained on a dataset including example images of slides and/or segments of images of slides, associated with whether a sample is present. Such a dataset may be gathered by, for example, capturing images of slides, and manually identifying ones containing a sample. In some embodiments, multiple machine learning models may be trained to identify different kinds of sample. Once sample identification machine learning model 1960 is trained, it may accept as an input an image of a region of interest and may output a determination as to whether a sample is present. In some embodiments, sample identification machine learning model 1960 may be improved, such as through use of reinforcement learning. Feedback used to determine a cost function of a reinforcement learning model may include, for example, user inputs, and annotations on a slide. For example, if an annotation, as transcribed using optical character recognition, indicates a particular type of sample, and sample identification machine learning model 1960 indicated that no regions of interest contain a sample, then a cost function indicating that the output is false may be determined. In another example, a user may input a label to be associated with a region of interest. If the label indicates a particular type of sample, and sample identification machine learning model 1960 output indicated that a sample was present in the region of interest, then a cost function may be determined indicating that the output was correct. Still referring to FIG.19, in some embodiments, whether a sample is present may be determined locally. For example, apparatus 1900 may include an already trained sample identification machine learning model 1960 and may apply the model to an image or segment of an image. In some embodiments, whether a sample is present may be determined externally. For example, apparatus 1900 may transmit image data to another computing device and may receive 116 Attorney Docket No.1519-029PCT1 a determination as to whether a sample is present. In some embodiments, whether a sample is present may be determined in real time. Still referring to FIG.19, in some embodiments, a machine vision system and/or an optical character recognition system may be used to determine one or more features of sample and/or slide 1916. In a non-limiting example, an optical character recognition system may be used to identify writing on slide 1916, and this may be used to annotate an image of slide 1916. Still referring to FIG.19, in some embodiments, apparatus 1900 may capture a plurality of images at differing sample focus depths. As used herein, a “sample focus depth” is a depth within a sample that an optical system is in focus. As used herein, a “focus distance” is an object side focal length. In some embodiments, first image and second image may have different focus distances and/or sample focus depths. Still referring to FIG.19, in some embodiments, apparatus 1900 may include a machine vision system. In some embodiments, a machine vision system may include at least a camera. A machine vision system may use images, such as images from at least a camera, to make a determination about a scene, space, and/or object. For example, in some cases a machine vision system may be used for world modeling or registration of objects within a space. In some cases, registration may include image processing, such as without limitation object recognition, feature detection, edge/corner detection, and the like. Non-limiting example of feature detection may include scale invariant feature transform (SIFT), Canny edge detection, Shi Tomasi corner detection, and the like. In some cases, registration may include one or more transformations to orient a camera frame (or an image or video stream) relative a three-dimensional coordinate system; exemplary transformations include without limitation homography transforms and affine transforms. In an embodiment, registration of first frame to a coordinate system may be verified and/or corrected using object identification and/or computer vision, as described above. For instance, and without limitation, an initial registration to two dimensions, represented for instance as registration to the x and y coordinates, may be performed using a two-dimensional projection of points in three dimensions onto a first frame, however. A third dimension of registration, representing depth and/or a z axis, may be detected by comparison of two frames; for instance, where first frame includes a pair of frames captured using a pair of cameras (e.g., stereoscopic camera also referred to in this disclosure as stereo-camera), image recognition and/or edge detection software may be used to detect a pair of stereoscopic views of images of 117 Attorney Docket No.1519-029PCT1 an object; two stereoscopic views may be compared to derive z-axis values of points on object permitting, for instance, derivation of further z-axis points within and/or around the object using interpolation. This may be repeated with multiple objects in field of view, including without limitation environmental features of interest identified by object classifier and/or indicated by an operator. In an embodiment, x and y axes may be chosen to span a plane common to two cameras used for stereoscopic image capturing and/or an xy plane of a first frame; a result, x and y translational components and ϕ may be pre-populated in translational and rotational matrices, for affine transformation of coordinates of object, also as described above. Initial x and y coordinates and/or guesses at transformational matrices may alternatively or additionally be performed between first frame and second frame, as described above. For each point of a plurality of points on object and/or edge and/or edges of object as described above, x and y coordinates of a first stereoscopic frame may be populated, with an initial estimate of z coordinates based, for instance, on assumptions about object, such as an assumption that ground is substantially parallel to an xy plane as selected above. Z coordinates, and/or x, y, and z coordinates, registered using image capturing and/or object identification processes as described above may then be compared to coordinates predicted using initial guess at transformation matrices; an error function may be computed using by comparing the two sets of points, and new x, y, and/or z coordinates, may be iteratively estimated and compared until the error function drops below a threshold level. In some cases, a machine vision system may use a classifier, such as any classifier described throughout this disclosure. Still referring to FIG.19, in some embodiments, image data may be processed using optical character recognition. In some embodiments, optical character recognition or optical character reader (OCR) includes automatic conversion of images of written (e.g., typed, handwritten or printed text) into machine-encoded text. In some cases, recognition of at least a keyword from image data may include one or more processes, including without limitation optical character recognition (OCR), optical word recognition, intelligent character recognition, intelligent word recognition, and the like. In some cases, OCR may recognize written text, one glyph or character at a time. In some cases, optical word recognition may recognize written text, one word at a time, for example, for languages that use a space as a word divider. In some cases, intelligent character recognition (ICR) may recognize written text one glyph or character at a time, for instance by employing machine-learning processes. In some cases, intelligent word 118 Attorney Docket No.1519-029PCT1 recognition (IWR) may recognize written text, one word at a time, for instance by employing machine-learning processes. Still referring to FIG.19, in some cases OCR may be an "offline" process, which analyses a static document or image frame. In some cases, handwriting movement analysis can be used as input to handwriting recognition. For example, instead of merely using shapes of glyphs and words, this technique may capture motions, such as the order in which segments are drawn, the direction, and the pattern of putting the pen down and lifting it. This additional information may make handwriting recognition more accurate. In some cases, this technology may be referred to as “online” character recognition, dynamic character recognition, real-time character recognition, and intelligent character recognition. Still referring to FIG.19, in some cases, OCR processes may employ pre-processing of image data. Pre-processing process may include without limitation de-skew, de-speckle, binarization, line removal, layout analysis or “zoning,” line and word detection, script recognition, character isolation or “segmentation,” and normalization. In some cases, a de-skew process may include applying a transform (e.g., homography or affine transform) to image data to align text. In some cases, a de-speckle process may include removing positive and negative spots and/or smoothing edges. In some cases, a binarization process may include converting an image from color or greyscale to black-and-white (i.e., a binary image). Binarization may be performed as a simple way of separating text (or any other desired image component) from a background of image data. In some cases, binarization may be required for example if an employed OCR algorithm only works on binary images. In some cases, a line removal process may include removal of non-glyph or non-character imagery (e.g., boxes and lines). In some cases, a layout analysis or “zoning” process may identify columns, paragraphs, captions, and the like as distinct blocks. In some cases, a line and word detection process may establish a baseline for word and character shapes and separate words, if necessary. In some cases, a script recognition process may, for example in multilingual documents, identify script allowing an appropriate OCR algorithm to be selected. In some cases, a character isolation or “segmentation” process may separate signal characters, for example character-based OCR algorithms. In some cases, a normalization process may normalize aspect ratio and/or scale of image data. Still referring to FIG.19, in some embodiments an OCR process may include an OCR algorithm. Exemplary OCR algorithms include matrix matching process and/or feature extraction 119 Attorney Docket No.1519-029PCT1 processes. Matrix matching may involve comparing an image to a stored glyph on a pixel-by- pixel basis. In some case, matrix matching may also be known as “pattern matching,” “pattern recognition,” and/or “image correlation.” Matrix matching may rely on an input glyph being correctly isolated from the rest of image data. Matrix matching may also rely on a stored glyph being in a similar font and at a same scale as input glyph. Matrix matching may work best with typewritten text. Still referring to FIG.19, in some embodiments, an OCR process may include a feature extraction process. In some cases, feature extraction may decompose a glyph into at least a feature. Exemplary non-limiting features may include corners, edges, lines, closed loops, line direction, line intersections, and the like. In some cases, feature extraction may reduce dimensionality of representation and may make the recognition process computationally more efficient. In some cases, extracted feature may be compared with an abstract vector-like representation of a character, which might reduce to one or more glyph prototypes. General techniques of feature detection in computer vision are applicable to this type of OCR. In some embodiments, machine-learning processes like nearest neighbor classifiers (e.g., k-nearest neighbors algorithm) may be used to compare image features with stored glyph features and choose a nearest match. OCR may employ any machine-learning process described in this disclosure, for example machine-learning processes described with reference to other figures. Exemplary non-limiting OCR software includes Cuneiform and Tesseract. Cuneiform is a multi- language, open-source optical character recognition system originally developed by Cognitive Technologies of Moscow, Russia. Tesseract is free OCR software originally developed by Hewlett-Packard of Palo Alto, California, United States. Still referring to FIG.19, in some cases, OCR may employ a two-pass approach to character recognition. A first pass may try to recognize a character. Each character that is satisfactory is passed to an adaptive classifier as training data. The adaptive classifier then gets a chance to recognize characters more accurately as it further analyzes image data. Since the adaptive classifier may have learned something useful a little too late to recognize characters on the first pass, a second pass is run over the image data. Second pass may include adaptive recognition and use characters recognized with high confidence on the first pass to recognize better remaining characters on the second pass. In some cases, two-pass approach may be advantageous for unusual fonts or low-quality image data. Another exemplary OCR software 120 Attorney Docket No.1519-029PCT1 tool include OCRopus. OCRopus development is led by German Research Centre for Artificial Intelligence in Kaiserslautern, Germany. In some cases, OCR software may employ neural networks. Still referring to FIG.19, in some cases, OCR may include post-processing. For example, OCR accuracy may be increased, in some cases, if output is constrained by a lexicon. A lexicon may include a list or set of words that are allowed to occur in a document. In some cases, a lexicon may include, for instance, all the words in the English language, or a more technical lexicon for a specific field. In some cases, an output stream may be a plain text stream or file of characters. In some cases, an OCR process may preserve an original layout of image data. In some cases, near-neighbor analysis can make use of co-occurrence frequencies to correct errors, by noting that certain words are often seen together. For example, “Washington, D.C.” is generally far more common in English than “Washington DOC.” In some cases, an OCR process may make us of a priori knowledge of grammar for a language being recognized. For example, grammar rules may be used to help determine if a word is likely to be a verb or a noun. Distance conceptualization may be employed for recognition and classification. For example, a Levenshtein distance algorithm may be used in OCR post-processing to further optimize results. Still referring to FIG.19, in some embodiments, apparatus 1900 may remove an artifact from an image. As used herein, an “artifact” is a visual inaccuracy, an element of an image that distracts from an element of interest, an element of an image that obscures an element of interest, or another undesirable element of an image. Still referring to FIG.19, apparatus 1900 may include an image processing module. As used in this disclosure, an “image processing module” is a component designed to process digital images. In an embodiment, image processing module may include a plurality of software algorithms that can analyze, manipulate, or otherwise enhance an image, such as, without limitation, a plurality of image processing techniques as described below. In another embodiment, image processing module may include hardware components such as, without limitation, one or more graphics processing units (GPUs) that can accelerate the processing of large amount of images. In some cases, image processing module may be implemented with one or more image processing libraries such as, without limitation, OpenCV, PIL/Pillow, ImageMagick, and the like. 121 Attorney Docket No.1519-029PCT1 Still referring to FIG.19, image processing module may be configured to receive images from optical sensor 1920. One or more images may be transmitted, from optical sensor 1920 to image processing module, via any suitable electronic communication protocol, including without limitation packet-based protocols such as transfer control protocol-internet protocol (TCP-IP), file transfer protocol (FTP) or the like. Receiving images may include retrieval of images from a data store containing images as described below; for instance, and without limitation, images may be retrieved using a query that specifies a timestamp that images may be required to match. Still referring to FIG.19, image processing module may be configured to process images. In an embodiment, image processing module may be configured to compress and/or encode images to reduce the file size and storage requirements while maintaining the essential visual information needed for further processing steps as described below. In an embodiment, compression and/or encoding of an image may facilitate faster transmission of images. In some cases, image processing module may be configured to perform a lossless compression on images, wherein the lossless compression may maintain the original image quality of images. In a non- limiting example, image processing module may utilize one or more lossless compression algorithms, such as, without limitation, Huffman coding, Lempel-Ziv-Welch (LZW), Run- Length Encoding (RLE), and/or the like to identify and remove redundancy in images without losing any information. In such embodiment, compressing and/or encoding each image of images may include converting the file format of each image into PNG, GIF, lossless JPEG2000 or the like. In an embodiment, images compressed via lossless compression may be perfectly reconstructed to the original form (e.g., original image resolution, dimension, color representation, format, and the like) of images. In other cases, image processing module may be configured to perform a lossy compression on images, wherein the lossy compression may sacrifice some image quality of images to achieve higher compression ratios. In a non-limiting example, image processing module may utilize one or more lossy compression algorithms, such as, without limitation, Discrete Cosine Transform (DCT) in JPEG or Wavelet Transform in JPEG2000, discard some less significant information within images, resulting in a smaller file size but a slight loss of image quality of images. In such embodiment, compressing and/or encoding images may include converting the file format of each image into JPEG, WebP, lossy JPEG2000, or the like. 122 Attorney Docket No.1519-029PCT1 Still referring to FIG.19, in an embodiment, processing images may include determining a degree of quality of depiction of a region of interest of an image. In an embodiment, image processing module may determine a degree of blurriness of images. In a non-limiting example, image processing module may perform a blur detection by taking a Fourier transform, or an approximation such as a Fast Fourier Transform (FFT) of images and analyzing a distribution of low and high frequencies in the resulting frequency-domain depiction of images; for instance, and without limitation, numbers of high-frequency values below a threshold level may indicate blurriness. In another non-limiting example, detection of blurriness may be performed by convolving images, a channel of images, or the like with a Laplacian kernel; for instance, and without limitation, this may generate a numerical score reflecting a number of rapid changes in intensity shown in each image, such that a high score indicates clarity and a low score indicates blurriness. In some cases, blurriness detection may be performed using a Gradient-based operator, which measures operators based on the gradient or first derivative of images, based on the hypothesis that rapid changes indicate sharp edges in the image, and thus are indicative of a lower degree of blurriness. In some cases, blur detection may be performed using Wavelet-based operator, which takes advantage of the capability of coefficients of the discrete wavelet transform to describe the frequency and spatial content of images. In some cases, blur detection may be performed using statistics-based operators take advantage of several image statistics as texture descriptors in order to compute a focus level. In other cases, blur detection may be performed by using discrete cosine transform (DCT) coefficients in order to compute a focus level of images from its frequency content. Additionally, or alternatively, image processing module may be configured to rank images according to degree of quality of depiction of a region of interest and select a highest-ranking image from a plurality of images. Still referring to FIG.19, processing images may include enhancing an image or at least a region of interest via a plurality of image processing techniques to improve the quality (or degree of quality of depiction) of an image for better processing and analysis as described further in this disclosure. In an embodiment, image processing module may be configured to perform a noise reduction operation on an image, wherein the noise reduction operation may remove or minimize noise (arises from various sources, such as sensor limitations, poor lighting conditions, image compression, and/or the like), resulting in a cleaner and more visually coherent image. In some cases, noise reduction operation may be performed using one or more image filters; for instance, 123 Attorney Docket No.1519-029PCT1 and without limitation, noise reduction operation may include Gaussian filtering, median filtering, bilateral filtering, and/or the like. Noise reduction operation may be done by image processing module, by averaging or filtering out pixel values in neighborhood of each pixel of an image to reduce random variations. Still referring to FIG.19, in another embodiment, image processing module may be configured to perform a contrast enhancement operation on an image. In some cases, an image may exhibit low contrast, which may, for example, make a feature difficult to distinguish from the background. Contrast enhancement operation may improve the contrast of an image by stretching the intensity range of the image and/or redistributing the intensity values (i.e., degree of brightness or darkness of a pixel in the image). In a non-limiting example, intensity value may represent the gray level or color of each pixel, scale from 0 to 255 in intensity range for an 8-bit image, and scale from 0 to 16,777,215 in a 24-bit color image. In some cases, contrast enhancement operation may include, without limitation, histogram equalization, adaptive histogram equalization (CLAHE), contrast stretching, and/or the like. Image processing module may be configured to adjust the brightness and darkness levels within an image to make a feature more distinguishable (i.e., increase degree of quality of depiction). Additionally, or alternatively, image processing module may be configured to perform a brightness normalization operation to correct variations in lighting conditions (i.e., uneven brightness levels). In some cases, an image may include a consistent brightness level across a region after brightness normalization operation performed by image processing module. In a non-limiting example, image processing module may perform a global or local mean normalization, where the average intensity value of an entire image or region of an image may be calculated and used to adjust the brightness levels. Still referring to FIG.19, in other embodiments, image processing module may be configured to perform a color space conversion operation to increase degree of quality of depiction. In a non-limiting example, in case of a color image (i.e., RGB image), image processing module may be configured to convert RGB image to grayscale or HSV color space. Such conversion may emphasize the differences in intensity values between a region or feature of interest and the background. Image processing module may further be configured to perform an image sharpening operation such as, without limitation, unsharp masking, Laplacian sharpening, high-pass filtering, and/or the like. Image processing module may use image 124 Attorney Docket No.1519-029PCT1 sharpening operation to enhance the edges and fine details related to a region or feature of interest within an image by emphasizing high-frequency components within an image. Still referring to FIG.19, processing images may include isolating a region or feature of interest from the rest of an image as a function of plurality of image processing techniques. Images may include highest-ranking image selected by image processing module as described above. In an embodiment, plurality of image processing techniques may include one or more morphological operations, wherein the morphological operations are techniques developed based on set theory, lattice theory, topology, and random functions used for processing geometrical structures using a structuring element. A “structuring element,” for the purpose of this disclosure, is a small matrix or kernel that defines a shape and size of a morphological operation. In some cases, structing element may be centered at each pixel of an image and used to determine an output pixel value for that location. In a non-limiting example, isolating a region or feature of interest from an image may include applying a dilation operation, wherein the dilation operation is a basic morphological operation configured to expand or grow the boundaries of objects (e.g., a cell, a dust particle, and the like) in an image. In another non-limiting example, isolating a region or feature of interest from an image may include applying an erosion operation, wherein the erosion operation is a basic morphological operation configured to shrink or erode the boundaries of objects in an image. In another non-limiting example, isolating a region or feature of interest from an image may include applying an opening operation, wherein the opening operation is a basic morphological operation configured to remove small objects or thin structures from an image while preserving larger structures. In a further non-limiting example, isolating a region or feature of interest from an image may include applying a closing operation, wherein the closing operation is a basic morphological operation configured to fill in small gaps or holes in objects in an image while preserving the overall shape and size of the objects. These morphological operations may be performed by image processing module to enhance the edges of objects, remove noise, or fill gaps in a region or feature of interest before further processing. Still referring to FIG.19, in an embodiment, isolating a region or feature of interest from an image may include utilizing an edge detection technique, which may detect one or more shapes defined by edges. An “edge detection technique,” as used in this disclosure, includes a mathematical method that identifies points in a digital image, at which the image brightness changes sharply and/or has a discontinuity. In an embodiment, such points may be organized into 125 Attorney Docket No.1519-029PCT1 straight and/or curved line segments, which may be referred to as “edges.” Edge detection technique may be performed by image processing module, using any suitable edge detection algorithm, including without limitation Canny edge detection, Sobel operator edge detection, Prewitt operator edge detection, Laplacian operator edge detection, and/or Differential edge detection. Edge detection technique may include phase congruency-based edge detection, which finds all locations of an image where all sinusoids in the frequency domain, for instance as generated using a Fourier decomposition, may have matching phases which may indicate a location of an edge. Edge detection technique may be used to detect a shape of a feature of interest such as a cell, indicating a cell membrane or wall; in an embodiment, edge detection technique may be used to find closed figures formed by edges. Still referring to FIG.19, in a non-limiting example, isolating a feature of interest from an image may include determining a feature of interest via edge detection technique. A feature of interest may include a specific area within a digital image that contains information relevant to further processing as described below. In a non-limiting example, an image data located outside a feature of interest may include irrelevant or extraneous information. Such portion of an image containing irrelevant or extraneous information may be disregarded by image processing module, thereby allowing resources to be concentrated at a feature of interest. In some cases, feature of interest may vary in size, shape, and/or location within an image. In a non-limiting example feature of interest may be presented as a circle around the nucleus of a cell. In some cases, feature of interest may specify one or more coordinates, distances and the like, such as center and radius of a circle around the nucleus of a cell in an image. Image processing module may then be configured to isolate feature of interest from the image based on feature of interest. In a non- limiting example, image processing module may crop an image according to a bounding box around a feature of interest. Still referring to FIG.19, image processing module may be configured to perform a connected component analysis (CCA) on an image for feature of interest isolation. As used in this disclosure, a “connected component analysis (CCA),” also known as connected component labeling, is an image processing technique used to identify and label connected regions within a binary image (i.e., an image which each pixel having only two possible values: 0 or 1, black or white, or foreground and background). “Connected regions,” as described herein, is a group of adjacent pixels that share the same value and are connected based on a predefined neighborhood 126 Attorney Docket No.1519-029PCT1 system such as, without limitation, 4-connected or 8-connected neighborhoods. In some cases, image processing module may convert an image into a binary image via a thresholding process, wherein the thresholding process may involve setting a threshold value that separates the pixels of an image corresponding to feature of interest (foreground) from those corresponding to the background. Pixels with intensity values above the threshold may be set to 1 (white) and those below the threshold may be set to 0 (black). In an embodiment, CCA may be employed to detect and extract feature of interest by identifying a plurality of connected regions that exhibit specific properties or characteristics of the feature of interest. Image processing module may then filter plurality of connected regions by analyzing plurality of connected regions properties such as, without limitation, area, aspect ratio, height, width, perimeter, and/or the like. In a non-limiting example, connected components that closely resemble the dimensions and aspect ratio of feature of interest may be retained, by image processing module as feature of interest, while other components may be discarded. Image processing module may be further configured to extract feature of interest from an image for further processing as described below. Still referring to FIG.19, in an embodiment, isolating feature of interest from an image may include segmenting a region depicting a feature of interest into a plurality sub-regions. Segmenting a region into sub-regions may include segmenting a region as a function of feature of interest and/or CCA via an image segmentation process. As used in this disclosure, an “image segmentation process” is a process for partition a digital image into one or more segments, where each segment represents a distinct part of the image. Image segmentation process may change the representation of images. Image segmentation process may be performed by image processing module. In a non-limiting example, image processing module may perform a region- based segmentation, wherein the region-based segmentation involves growing regions from one or more seed points or pixels on an image based on a similarity criterion. Similarity criterion may include, without limitation, color, intensity, texture, and/or the like. In a non-limiting example, region-based segmentation may include region growing, region merging, watershed algorithms, and the like. Still referring to FIG.19, in some embodiments, apparatus 1900 may remove an artifact identified by a machine vision system or an optical character recognition system, which are described above. Non-limiting examples of artifacts that may be removed include dust particles, bubbles, cracks in slide 1916, writing on slide 1916, shadows, visual noise such as in a grainy 127 Attorney Docket No.1519-029PCT1 image, and the like. In some embodiments, an artifact may be partially removed and/or lowered in visibility. Still referring to FIG.19, in some embodiments, an artifact may be removed using an artifact removal machine learning model. In some embodiments, artifact removal machine learning model may be trained on a dataset including images, associated with images without artifacts. In some embodiments, artifact removal machine learning model may accept as an input an image including an artifact and may output an image without the artifact. For example, artifact removal machine learning model may accept as an input an image including a bubble in a slide and may output an image that does not include the bubble. In some embodiments, artifact removal machine learning model may include a generative machine learning model such as a diffusion model. A diffusion model may learn the structure of a dataset by modeling the way data points diffuse through a latent space. In some embodiments, artifact removal may be done locally. For example, apparatus 1900 may include an already trained artifact removal machine learning model and may apply the model to an image. In some embodiments, artifact removal may be done externally. For example, apparatus 1900 may transmit image data to another computing device and may receive an image with an artifact removed. In some embodiments, an artifact may be removed in real time. In some embodiments, an artifact may be removed based on identification by a user. For example, a user may drag a box around an artifact using a mouse cursor, and apparatus 1900 may remove an artifact in the box. Still referring to FIG.19, in some embodiments, apparatus 1900 may display an image to a user. In some embodiments, first image may be displayed to a user in real time. In some embodiments, an image may be displayed to a user using output interface 1932. For example, first image may be displayed to a user a display such as a screen. In some embodiments, first image may be displayed to a user in the context of a graphical user interface (GUI). For example, a GUI may include controls for navigating an image such as controls for zooming in or out or changing where is being viewed. A GUI may include a touchscreen. Still referring to FIG.19, in some embodiments, apparatus 1900 may receive a parameter set from a user. As used herein, a “parameter set” is a set of values that identify how an image is to be captured. A parameter set may be implemented as a data structure as described below. In some embodiments, apparatus 1900 may receive a parameter set from a user using input interface 1928. A parameter set may include X and Y coordinates indicating where the user 128 Attorney Docket No.1519-029PCT1 wishes to view. A parameter set may include a desired level of magnification. As used herein, a “level of magnification” is a datum describing how zoomed in or out an image is to be captured at. A level of magnification may account for optical zoom and/or digital zoom. As a non-limiting example, a level of magnification may be “8x” magnification. A parameter set may include a desired sample focus depth and/or focus distance. In a non-limiting example, user may manipulate input interface 1928 such that a parameter set includes X and Y coordinates and a level of magnification corresponding to a more zoomed in view of a particular region of a sample. In some embodiments, a parameter set corresponds to a more zoomed in view of a particular region of a sample that is included in first image. This may be done, for example, to get a more detailed view of a small object. As used herein, unless indicated otherwise, an “X coordinate” and a “Y coordinate” refer to coordinates along perpendicular axes, where the plane defined by these axes is parallel to a plane of a surface of slide 1916. In some cases, setting a magnification may include changing one or more optical elements within optical system. For example, setting a magnification may include replacing a first objective lens for a second objective lens having a different magnification. Additionally or alternative, one or more optical components "down beam" from objective lens may be replaced to change a total magnification of optical system and, thereby, set a magnification. In some cases, setting a magnification may include changing a digital magnification. Digital magnification may include outputting an image, using output interface, at a different resolution, i.e. after re-scaling the image. In some embodiments, apparatus 1900 may capture an image with a particular field of view. A field of view may be determined based on, for example, a level of magnification, and how wide of an angle a camera captures images at. Still referring to FIG.19, in some embodiments, apparatus 1900 may move one or more of slide port 1940, slide 1916, and at least an optical system into a second position. In some embodiments, the location of a second position may be based on a parameter set. In some embodiments, the location of a second position may be based on an identification of a region of interest, row, or point, as described above. Second position may be determined by, for example, modifying the position of optical system relative to slide 1916 based on parameter set. For example, parameter set may indicate that second position is achieved by modifying an X coordinate by 5mm in a particular direction. In this example, second position may be found by modifying optical system’s original position by 5mm in that direction. In some embodiments, 129 Attorney Docket No.1519-029PCT1 such movement may be done using actuator mechanism 1924. In some embodiments, actuator mechanism 1924 may move slide port 1940 such that slide 1916 is in a position relative to at least an optical system such that optical sensor 1920 may capture an image as directed by parameter set. For example, slide 1916 may rest on slide port 1940 and movement of slide port 1940 may move slide 1916 as well. In some embodiments, actuator mechanism 1924 may move slide 1916 such that slide 1916 is in a position relative to at least an optical system such that optical sensor 1920 may capture an image as directed by parameter set. For example, slide 1916 may be connected to actuator mechanism 1924 such that actuator mechanism 1924 may move slide 1916 relative to at least an optical system. In some embodiments, actuator mechanism 1924 may move at least an optical system such that slide 1916 is in a position relative to slide 1916 such that optical system may capture an image as directed by parameter set. For example, slide 1916 may be stationary, and actuator mechanism 1924 may move at least an optical system into position relative to slide 1916. In some embodiments, actuator mechanism 1924 may move more than one of slide port 1940, slide 1916, and at least an optical system such that they are in the correct relative positions. In some embodiments, actuator mechanism 1924 may move slide port 1940, slide 1916, and/or at least an optical system in real time. For example, user input of a parameter set may cause a substantially immediate movement of items by actuator mechanism 1924. Still referring to FIG.19, in some embodiments, apparatus 1900 may capture a second image of slide 1916 at second position. In some embodiments, apparatus 1900 may capture second image using at least an optical system. In some embodiments, second image may include an image of a region of a sample. In some embodiments, second image may include an image of a region of an area captured in first image. For example, second image may include a more zoomed in, higher resolution per unit area, image of a region within first image. In another example, second image may be captured using a focus distance based on a focus pattern. This may cause display of second image to allow a user to detect smaller details within the imaged region. Still referring to FIG.19, in some embodiments, second image includes a shift in X and Y coordinates relative to first image. For example, second image may partially overlap with first image. 130 Attorney Docket No.1519-029PCT1 Still referring to FIG.19, in some embodiments, apparatus 1900 may capture second image in real time. For example, user may manipulate input interface 1928, creating parameter set, then actuator mechanism 1924 may cause movement of slide 1916 relative to optical system to start substantially immediately after the input interface 1928 was manipulated, then optical system may capture second image substantially immediately after actuator mechanism 1924 completes its movement. In some embodiments, artifacts may also be removed in real time. In some embodiments, image may also be annotated in real time. In some embodiments, focus pattern may also be determined in real time and images taken after first image may be taken using a focus distance according to focus pattern. Still referring to FIG.19, in some embodiments, apparatus 1900 may display second image to user. In some embodiments, second image may be displayed to user using output device. Second image may be displayed as described above with respect to output device and display of first image. In some embodiments, displaying second image to user may include replacing a region of first image with second image to create a hybrid image; and displaying the hybrid image to the user. As used herein, a “hybrid image” is an image constructed by combining a first image with a second image. In some embodiments, creation of a hybrid image in this way may preserve second image. For example, if second image covers a smaller area at a higher resolution per unit area than first image, then second image may replace a lower resolution per unit area segment of first area corresponding to the region covered by second image. In some embodiments, image adjustments may be made to offset visual differences between first image and second image at a border of first image and second image in hybrid image. In a non-limiting example, color may be adjusted such that background color of the images is consistent along the border of the images. As another non-limiting example, brightness of the images may be adjusted such that there is no stark difference between brightness of the images. In some embodiments, artifacts may be removed from second image and/or hybrid image as described above. In some embodiments, second image may be displayed to user in real time. For example, adjustments (such as annotations and/or artifact removal) may be started substantially immediately after second image is captured, and an adjusted version of second image may be displayed to user substantially immediately after adjustments are done. In some embodiments, an unadjusted version of second image may be displayed to user while adjustments are being made. In some 131 Attorney Docket No.1519-029PCT1 embodiments, where there are a plurality of images covering a specific region, when user zooms out using user interface 1936, a lower resolution image of the region may be displayed. Still referring to FIG.19, in some embodiments, apparatus 1900 may transmit first image, second image, hybrid image, and/or a data structure including a plurality of images to an external device. Such an external device may include, in non-limiting examples, a phone, tablet, or computer. In some embodiments, such a transmission may configure the external device to display an image. Still referring to FIG.19, in some embodiments, apparatus 1900 may annotate an image. In some embodiments, apparatus 1900 may annotate first image. In some embodiments, apparatus 1900 may annotate second image. In some embodiments, apparatus 1900 may annotate hybrid image. For example, upon creation of hybrid image, apparatus 1900 may recognize a cell depicted in hybrid image as a cell of a particular type and may annotate hybrid image indicating the cell type. In a non-limiting example, apparatus 1900 may associate text with a particular location in an image, where the text describes a feature present at that location in the image. In some embodiments, apparatus 1900 may annotate an image selected from the list consisting of first image, second image and hybrid image. Still referring to FIG.19, in some embodiments, annotations may be made as a function of user input of an annotation instruction into input interface 1928. As used herein, an “annotation instruction” is a datum generated based on user input indicating whether to create an annotation or describing an annotation to be made. For example, user may select an option controlling whether apparatus 1900 annotates images. In another example, user may manually annotate an image. In some embodiments, annotations may be made automatically. In some embodiments, annotations may be made using an annotation machine learning model. In some embodiments, annotation machine learning model may include an optical character recognition model, as described above. In some embodiments, annotation machine learning model may be trained using a dataset including image data, associated with text depicted by the image. In some embodiments, annotation machine learning model may accept as an input image data and may output annotated image data and/or annotations to apply to the image data. In some embodiments, annotation machine learning model may be used to convert text written on slides 1916 to annotations on images. In some embodiments, annotation machine learning model may include a machine vision model, as described above. In some embodiments, annotation machine 132 Attorney Docket No.1519-029PCT1 learning model including a machine vision model may be trained on a data set including image data, associated with annotations indicating features of the image data. In some embodiments, annotation machine learning model may accept as an input image data and may output annotated image data and/or annotations to apply to image data. Non-limiting examples of features annotation machine learning model may be trained to recognize include cell types, features of cells, and objects in a slide 1916 such as bubbles. In some embodiments, images may be annotated in real time. For example, annotation may start substantially immediately after an image is captured and/or a command to annotate an image is received, and annotated image may be displayed to user substantially immediately after annotation is completed. Still referring to FIG.19, in some embodiments, apparatus 1900 may determine a visual element data structure. In some embodiments, apparatus 1900 may display to a user a visual element as a function of visual element data structure. As used herein, a “visual element data structure” is a data structure describing a visual element. As non-limiting examples, visual elements may include first image, second image, hybrid image, and elements of a GUI. Still referring to FIG.19, in some embodiments, a visual element data structure may include a visual element. As used herein, a “visual element” is a datum that is displayed visually to a user. In some embodiments, a visual element data structure may include a rule for displaying visual element. In some embodiments, a visual element data structure may be determined as a function of first image, second image, and/or hybrid image. In some embodiments, a visual element data structure may be determined as a function of an item from the list consisting of first image, second image, hybrid image, a GUI element, and an annotation. In a non-limiting example, a visual element data structure may be generated such that visual element describing or a feature of first image, such as an annotation, is displayed to a user. Still referring to FIG.19, in some embodiments, visual element may include one or more elements of text, images, shapes, charts, particle effects, interactable features, and the like. As a non-limiting example, a visual element may include a touch screen button for setting magnification level. Still referring to FIG.19, a visual element data structure may include rules governing if or when visual element is displayed. In a non-limiting example, a visual element data structure may include a rule causing a visual element including an annotation describing first image, 133 Attorney Docket No.1519-029PCT1 second image, and/or hybrid image to be displayed when a user selects a specific region of first image, second image, and/or hybrid image using a GUI. Still referring to FIG.19, a visual element data structure may include rules for presenting more than one visual element, or more than one visual element at a time. In an embodiment, about 1, 2, 3, 4, 5, 10, 20, or 50 visual elements are displayed simultaneously. For example, a plurality of annotations may be displayed simultaneously. Still referring to FIG.19, in some embodiments, apparatus 1900 may transmit visual element to a display such as output interface 1932. A display may communicate visual element to user. A display may include, for example, a smartphone screen, a computer screen, or a tablet screen. A display may be configured to provide a visual interface. A visual interface may include one or more virtual interactive elements such as, without limitation, buttons, menus, and the like. A display may include one or more physical interactive elements, such as buttons, a computer mouse, or a touchscreen, that allow user to input data into the display. Interactive elements may be configured to enable interaction between a user and a computing device. In some embodiments, a visual element data structure is determined as a function of data input by user into a display. Still referring to FIG.19, a variable and/or datum described herein may be represented as a data structure. In some embodiments, a data structure may include one or more functions and/or variables, as a class might in object-oriented programming. In some embodiments, a data structure may include data in the form of a Boolean, integer, float, string, date, and the like. In a non-limiting example, an annotation data structure may include a string value representing text of the annotation. In some embodiments, data in a data structure may be organized in a linked list, tree, array, matrix, tenser, and the like. In a non-limiting example, annotation data structures may be organized in an array. In some embodiments, a data structure may include or be associated with one or more elements of metadata. A data structure may include one or more self-referencing data elements, which processor 1904 may use in interpreting the data structure. In a non-limiting example, a data structure may include “<date>” and “</date>,” tags, indicating that the content between the tags is a date. Still referring to FIG.19, a data structure may be stored in, for example, memory 1908 or a database. Database may be implemented, without limitation, as a relational database, a key- value retrieval database such as a NOSQL database, or any other format or structure for use as a 134 Attorney Docket No.1519-029PCT1 database that a person skilled in the art would recognize as suitable upon review of the entirety of this disclosure. Database may alternatively or additionally be implemented using a distributed data storage protocol and/or data structure, such as a distributed hash table or the like. Database may include a plurality of data entries and/or records as described above. Data entries in a database may be flagged with or linked to one or more additional elements of information, which may be reflected in data entry cells and/or in linked tables such as tables related by one or more indices in a relational database. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which data entries in a database may store, retrieve, organize, and/or reflect data and/or records as used herein, as well as categories and/or populations of data consistently with this disclosure. Still referring to FIG.19, in some embodiments, a data structure may be read and/or manipulated by processor 1904. In a non-limiting example, an image data structure may be read and displayed to user. In another non-limiting example, an image data structure may be modified to remove an artifact, as described above. Still referring to FIG.19, in some embodiments, a data structure may be calibrated. In some embodiments, a data structure may be trained using a machine learning algorithm. In a non-limiting example, a data structure may include an array of data representing the biases of connections of a neural network. In this example, the neural network may be trained on a set of training data, and a back propagation algorithm may be used to modify the data in the array. Machine learning models and neural networks are described further herein. Referring now to FIG.20, an exemplary embodiment of a machine-learning module 2000 that may perform one or more machine-learning processes as described in this disclosure is illustrated. Machine-learning module may perform determinations, classification, and/or analysis steps, methods, processes, or the like as described in this disclosure using machine learning processes. A “machine learning process,” as used in this disclosure, is a process that automatedly uses training data 2004 to generate an algorithm instantiated in hardware or software logic, data structures, and/or functions that will be performed by a computing device/module to produce outputs 2008 given data provided as inputs 2012; this is in contrast to a non-machine learning software program where the commands to be executed are determined in advance by a user and written in a programming language. 135 Attorney Docket No.1519-029PCT1 Still referring to FIG.20, “training data,” as used herein, is data containing correlations that a machine-learning process may use to model relationships between two or more categories of data elements. For instance, and without limitation, training data 2004 may include a plurality of data entries, also known as “training examples,” each entry representing a set of data elements that were recorded, received, and/or generated together; data elements may be correlated by shared existence in a given data entry, by proximity in a given data entry, or the like. Multiple data entries in training data 2004 may evince one or more trends in correlations between categories of data elements; for instance, and without limitation, a higher value of a first data element belonging to a first category of data element may tend to correlate to a higher value of a second data element belonging to a second category of data element, indicating a possible proportional or other mathematical relationship linking values belonging to the two categories. Multiple categories of data elements may be related in training data 2004 according to various correlations; correlations may indicate causative and/or predictive links between categories of data elements, which may be modeled as relationships such as mathematical relationships by machine-learning processes as described in further detail below. Training data 2004 may be formatted and/or organized by categories of data elements, for instance by associating data elements with one or more descriptors corresponding to categories of data elements. As a non- limiting example, training data 2004 may include data entered in standardized forms by persons or processes, such that entry of a given data element in a given field in a form may be mapped to one or more descriptors of categories. Elements in training data 2004 may be linked to descriptors of categories by tags, tokens, or other data elements; for instance, and without limitation, training data 2004 may be provided in fixed-length formats, formats linking positions of data to categories such as comma-separated value (CSV) formats and/or self-describing formats such as extensible markup language (XML), JavaScript Object Notation (JSON), or the like, enabling processes or devices to detect categories of data. Alternatively or additionally, and continuing to refer to FIG.20, training data 2004 may include one or more elements that are not categorized; that is, training data 2004 may not be formatted or contain descriptors for some elements of data. Machine-learning algorithms and/or other processes may sort training data 2004 according to one or more categorizations using, for instance, natural language processing algorithms, tokenization, detection of correlated values in raw data and the like; categories may be generated using correlation and/or other processing 136 Attorney Docket No.1519-029PCT1 algorithms. As a non-limiting example, in a corpus of text, phrases making up a number “n” of compound words, such as nouns modified by other nouns, may be identified according to a statistically significant prevalence of n-grams containing such words in a particular order; such an n-gram may be categorized as an element of language such as a “word” to be tracked similarly to single words, generating a new category as a result of statistical analysis. Similarly, in a data entry including some textual data, a person’s name may be identified by reference to a list, dictionary, or other compendium of terms, permitting ad-hoc categorization by machine- learning algorithms, and/or automated association of data in the data entry with descriptors or into a given format. The ability to categorize data entries automatedly may enable the same training data 2004 to be made applicable for two or more distinct machine-learning algorithms as described in further detail below. Training data 2004 used by machine-learning module 2000 may correlate any input data as described in this disclosure to any output data as described in this disclosure. As a non-limiting illustrative example, inputs may include image data and outputs may include text data. Further referring to FIG.20, training data may be filtered, sorted, and/or selected using one or more supervised and/or unsupervised machine-learning processes and/or models as described in further detail below; such models may include without limitation a training data classifier 2016. Training data classifier 2016 may include a “classifier,” which as used in this disclosure is a machine-learning model as defined below, such as a data structure representing and/or using a mathematical model, neural net, or program generated by a machine learning algorithm known as a “classification algorithm,” as described in further detail below, that sorts inputs into categories or bins of data, outputting the categories or bins of data and/or labels associated therewith. A classifier may be configured to output at least a datum that labels or otherwise identifies a set of data that are clustered together, found to be close under a distance metric as described below, or the like. A distance metric may include any norm, such as, without limitation, a Pythagorean norm. Machine-learning module 2000 may generate a classifier using a classification algorithm, defined as a processes whereby a computing device and/or any module and/or component operating thereon derives a classifier from training data 2004. Classification may be performed using, without limitation, linear classifiers such as without limitation logistic regression and/or naive Bayes classifiers, nearest neighbor classifiers such as k-nearest neighbors classifiers, support vector machines, least squares support vector machines, fisher’s linear 137 Attorney Docket No.1519-029PCT1 discriminant, quadratic classifiers, decision trees, boosted trees, random forest classifiers, learning vector quantization, and/or neural network-based classifiers. As a non-limiting example, training data classifier 2016 may classify elements of training data to individual glyphs as in an OCR machine learning model. With further reference to FIG.20, training examples for use as training data may be selected from a population of potential examples according to cohorts relevant to an analytical problem to be solved, a classification task, or the like. Alternatively or additionally, training data may be selected to span a set of likely circumstances or inputs for a machine-learning model and/or process to encounter when deployed. For instance, and without limitation, for each category of input data to a machine-learning process or model that may exist in a range of values in a population of phenomena such as images, user data, process data, physical data, or the like, a computing device, processor, and/or machine-learning model may select training examples representing each possible value on such a range and/or a representative sample of values on such a range. Selection of a representative sample may include selection of training examples in proportions matching a statistically determined and/or predicted distribution of such values according to relative frequency, such that, for instance, values encountered more frequently in a population of data so analyzed are represented by more training examples than values that are encountered less frequently. Alternatively or additionally, a set of training examples may be compared to a collection of representative values in a database and/or presented to a user, so that a process can detect, automatically or via user input, one or more values that are not included in the set of training examples. Computing device, processor, and/or module may automatically generate a missing training example; this may be done by receiving and/or retrieving a missing input and/or output value and correlating the missing input and/or output value with a corresponding output and/or input value collocated in a data record with the retrieved value, provided by a user and/or other device, or the like. Still referring to FIG.20, computer, processor, and/or module may be configured to sanitize training data. “Sanitizing” training data, as used in this disclosure, is a process whereby training examples are removed that interfere with convergence of a machine-learning model and/or process to a useful result. For instance, and without limitation, a training example may include an input and/or output value that is an outlier from typically encountered values, such that a machine-learning algorithm using the training example will be adapted to an unlikely 138 Attorney Docket No.1519-029PCT1 amount as an input and/or output; a value that is more than a threshold number of standard deviations away from an average, mean, or expected value, for instance, may be eliminated. Alternatively or additionally, one or more training examples may be identified as having poor quality data, where “poor quality” is defined as having a signal to noise ratio below a threshold value. As a non-limiting example, and with further reference to FIG.20, images used to train an image classifier or other machine-learning model and/or process that takes images as inputs or generates images as outputs may be rejected if image quality is below a threshold value. For instance, and without limitation, computing device, processor, and/or module may perform blur detection, and eliminate one or more Blur detection may be performed, as a non-limiting example, by taking Fourier transform, or an approximation such as a Fast Fourier Transform (FFT) of the image and analyzing a distribution of low and high frequencies in the resulting frequency-domain depiction of the image; numbers of high-frequency values below a threshold level may indicate blurriness. As a further non-limiting example, detection of blurriness may be performed by convolving an image, a channel of an image, or the like with a Laplacian kernel; this may generate a numerical score reflecting a number of rapid changes in intensity shown in the image, such that a high score indicates clarity and a low score indicates blurriness. Blurriness detection may be performed using a gradient-based operator, which measures operators based on the gradient or first derivative of an image, based on the hypothesis that rapid changes indicate sharp edges in the image, and thus are indicative of a lower degree of blurriness. Blur detection may be performed using Wavelet -based operator, which takes advantage of the capability of coefficients of the discrete wavelet transform to describe the frequency and spatial content of images. Blur detection may be performed using statistics-based operators take advantage of several image statistics as texture descriptors in order to compute a focus level. Blur detection may be performed by using discrete cosine transform (DCT) coefficients in order to compute a focus level of an image from its frequency content. Continuing to refer to FIG.20, computing device, processor, and/or module may be configured to precondition one or more training examples. For instance, and without limitation, where a machine learning model and/or process has one or more inputs and/or outputs requiring, transmitting, or receiving a certain number of bits, samples, or other units of data, one or more training examples’ elements to be used as or compared to inputs and/or outputs may be modified 139 Attorney Docket No.1519-029PCT1 to have such a number of units of data. For instance, a computing device, processor, and/or module may convert a smaller number of units, such as in a low pixel count image, into a desired number of units, for instance by upsampling and interpolating. As a non-limiting example, a low pixel count image may have 200 pixels, however a desired number of pixels may be 128. Processor may interpolate the low pixel count image to convert the 100 pixels into 128 pixels. It should also be noted that one of ordinary skill in the art, upon reading this disclosure, would know the various methods to interpolate a smaller number of data units such as samples, pixels, bits, or the like to a desired number of such units. In some instances, a set of interpolation rules may be trained by sets of highly detailed inputs and/or outputs and corresponding inputs and/or outputs downsampled to smaller numbers of units, and a neural network or other machine learning model that is trained to predict interpolated pixel values using the training data. As a non-limiting example, a sample input and/or output, such as a sample picture, with sample- expanded data units (e.g., pixels added between the original pixels) may be input to a neural network or machine-learning model and output a pseudo replica sample-picture with dummy values assigned to pixels between the original pixels based on a set of interpolation rules. As a non-limiting example, in the context of an image classifier, a machine-learning model may have a set of interpolation rules trained by sets of highly detailed images and images that have been downsampled to smaller numbers of pixels, and a neural network or other machine learning model that is trained using those examples to predict interpolated pixel values in a facial picture context. As a result, an input with sample-expanded data units (the ones added between the original data units, with dummy values) may be run through a trained neural network and/or model, which may fill in values to replace the dummy values. Alternatively or additionally, processor, computing device, and/or module may utilize sample expander methods, a low-pass filter, or both. As used in this disclosure, a “low-pass filter” is a filter that passes signals with a frequency lower than a selected cutoff frequency and attenuates signals with frequencies higher than the cutoff frequency. The exact frequency response of the filter depends on the filter design. Computing device, processor, and/or module may use averaging, such as luma or chroma averaging in images, to fill in data units in between original data units. In some embodiments, and with continued reference to FIG.20, computing device, processor, and/or module may down-sample elements of a training example to a desired lower number of data elements. As a non-limiting example, a high pixel count image may have 256 140 Attorney Docket No.1519-029PCT1 pixels, however a desired number of pixels may be 128. Processor may down-sample the high pixel count image to convert the 256 pixels into 128 pixels. In some embodiments, processor may be configured to perform downsampling on data. Downsampling, also known as decimation, may include removing every Nth entry in a sequence of samples, all but every Nth entry, or the like, which is a process known as “compression,” and may be performed, for instance by an N-sample compressor implemented using hardware or software. Anti-aliasing and/or anti-imaging filters, and/or low-pass filters, may be used to clean up side-effects of compression. Still referring to FIG.20, machine-learning module 2000 may be configured to perform a lazy-learning process 2020 and/or protocol, which may alternatively be referred to as a “lazy loading” or “call-when-needed” process and/or protocol, may be a process whereby machine learning is conducted upon receipt of an input to be converted to an output, by combining the input and training set to derive the algorithm to be used to produce the output on demand. For instance, an initial set of simulations may be performed to cover an initial heuristic and/or “first guess” at an output and/or relationship. As a non-limiting example, an initial heuristic may include a ranking of associations between inputs and elements of training data 2004. Heuristic may include selecting some number of highest-ranking associations and/or training data 2004 elements. Lazy learning may implement any suitable lazy learning algorithm, including without limitation a K-nearest neighbors algorithm, a lazy naïve Bayes algorithm, or the like; persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various lazy- learning algorithms that may be applied to generate outputs as described in this disclosure, including without limitation lazy learning applications of machine-learning algorithms as described in further detail below. Alternatively or additionally, and with continued reference to FIG.20, machine-learning processes as described in this disclosure may be used to generate machine-learning models 2024. A “machine-learning model,” as used in this disclosure, is a data structure representing and/or instantiating a mathematical and/or algorithmic representation of a relationship between inputs and outputs, as generated using any machine-learning process including without limitation any process as described above, and stored in memory; an input is submitted to a machine-learning model 2024 once created, which generates an output based on the relationship that was derived. For instance, and without limitation, a linear regression model, generated using a linear 141 Attorney Docket No.1519-029PCT1 regression algorithm, may compute a linear combination of input data using coefficients derived during machine-learning processes to calculate an output datum. As a further non-limiting example, a machine-learning model 2024 may be generated by creating an artificial neural network, such as a convolutional neural network comprising an input layer of nodes, one or more intermediate layers, and an output layer of nodes. Connections between nodes may be created via the process of "training" the network, in which elements from a training data 2004 set are applied to the input nodes, a suitable training algorithm (such as Levenberg-Marquardt, conjugate gradient, simulated annealing, or other algorithms) is then used to adjust the connections and weights between nodes in adjacent layers of the neural network to produce the desired values at the output nodes. This process is sometimes referred to as deep learning. Still referring to FIG.20, machine-learning algorithms may include at least a supervised machine-learning process 2028. At least a supervised machine-learning process 2028, as defined herein, include algorithms that receive a training set relating a number of inputs to a number of outputs, and seek to generate one or more data structures representing and/or instantiating one or more mathematical relations relating inputs to outputs, where each of the one or more mathematical relations is optimal according to some criterion specified to the algorithm using some scoring function. For instance, a supervised learning algorithm may include image data as described above as inputs, classification to particular characters as outputs, and a scoring function representing a desired form of relationship to be detected between inputs and outputs; scoring function may, for instance, seek to maximize the probability that a given input and/or combination of elements inputs is associated with a given output to minimize the probability that a given input is not associated with a given output. Scoring function may be expressed as a risk function representing an “expected loss” of an algorithm relating inputs to outputs, where loss is computed as an error function representing a degree to which a prediction generated by the relation is incorrect when compared to a given input-output pair provided in training data 2004. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various possible variations of at least a supervised machine-learning process 2028 that may be used to determine relation between inputs and outputs. Supervised machine-learning processes may include classification algorithms as defined above. With further reference to FIG.20, training a supervised machine-learning process may include, without limitation, iteratively updating coefficients, biases, weights based on an error 142 Attorney Docket No.1519-029PCT1 function, expected loss, and/or risk function. For instance, an output generated by a supervised machine-learning model using an input example in a training example may be compared to an output example from the training example; an error function may be generated based on the comparison, which may include any error function suitable for use with any machine-learning algorithm described in this disclosure, including a square of a difference between one or more sets of compared values or the like. Such an error function may be used in turn to update one or more weights, biases, coefficients, or other parameters of a machine-learning model through any suitable process including without limitation gradient descent processes, least-squares processes, and/or other processes described in this disclosure. This may be done iteratively and/or recursively to gradually tune such weights, biases, coefficients, or other parameters. Updating may be performed, in neural networks, using one or more back-propagation algorithms. Iterative and/or recursive updates to weights, biases, coefficients, or other parameters as described above may be performed until currently available training data is exhausted and/or until a convergence test is passed, where a “convergence test” is a test for a condition selected as indicating that a model and/or weights, biases, coefficients, or other parameters thereof has reached a degree of accuracy. A convergence test may, for instance, compare a difference between two or more successive errors or error function values, where differences below a threshold amount may be taken to indicate convergence. Alternatively or additionally, one or more errors and/or error function values evaluated in training iterations may be compared to a threshold. Still referring to FIG.20, a computing device, processor, and/or module may be configured to perform method, method step, sequence of method steps and/or algorithm described in reference to this figure, in any order and with any degree of repetition. For instance, a computing device, processor, and/or module may be configured to perform a single step, sequence and/or algorithm repeatedly until a desired or commanded outcome is achieved; repetition of a step or a sequence of steps may be performed iteratively and/or recursively using outputs of previous repetitions as inputs to subsequent repetitions, aggregating inputs and/or outputs of repetitions to produce an aggregate result, reduction or decrement of one or more variables such as global variables, and/or division of a larger processing task into a set of iteratively addressed smaller processing tasks. A computing device, processor, and/or module may perform any step, sequence of steps, or algorithm in parallel, such as simultaneously and/or substantially simultaneously performing a step two or more times using two or more parallel 143 Attorney Docket No.1519-029PCT1 threads, processor cores, or the like; division of tasks between parallel threads and/or processes may be performed according to any protocol suitable for division of tasks between iterations. Persons skilled in the art, upon reviewing the entirety of this disclosure, will be aware of various ways in which steps, sequences of steps, processing tasks, and/or data may be subdivided, shared, or otherwise dealt with using iteration, recursion, and/or parallel processing. Further referring to FIG.20, machine learning processes may include at least an unsupervised machine-learning processes 2032. An unsupervised machine-learning process, as used herein, is a process that derives inferences in datasets without regard to labels; as a result, an unsupervised machine-learning process may be free to discover any structure, relationship, and/or correlation provided in the data. Unsupervised processes 2032 may not require a response variable; unsupervised processes 2032 may be used to find interesting patterns and/or inferences between variables, to determine a degree of correlation between two or more variables, or the like. Still referring to FIG.20, machine-learning module 2000 may be designed and configured to create a machine-learning model 2024 using techniques for development of linear regression models. Linear regression models may include ordinary least squares regression, which aims to minimize the square of the difference between predicted outcomes and actual outcomes according to an appropriate norm for measuring such a difference (e.g. a vector-space distance norm); coefficients of the resulting linear equation may be modified to improve minimization. Linear regression models may include ridge regression methods, where the function to be minimized includes the least-squares function plus term multiplying the square of each coefficient by a scalar amount to penalize large coefficients. Linear regression models may include least absolute shrinkage and selection operator (LASSO) models, in which ridge regression is combined with multiplying the least-squares term by a factor of 1 divided by double the number of samples. Linear regression models may include a multi-task lasso model wherein the norm applied in the least-squares term of the lasso model is the Frobenius norm amounting to the square root of the sum of squares of all terms. Linear regression models may include the elastic net model, a multi-task elastic net model, a least angle regression model, a LARS lasso model, an orthogonal matching pursuit model, a Bayesian regression model, a logistic regression model, a stochastic gradient descent model, a perceptron model, a passive aggressive algorithm, a robustness regression model, a Huber regression model, or any other suitable model that may occur to persons skilled in the art upon reviewing the entirety of this disclosure. Linear 144 Attorney Docket No.1519-029PCT1 regression models may be generalized in an embodiment to polynomial regression models, whereby a polynomial equation (e.g. a quadratic, cubic or higher-order equation) providing a best predicted output/actual output fit is sought; similar methods to those described above may be applied to minimize error functions, as will be apparent to persons skilled in the art upon reviewing the entirety of this disclosure. Continuing to refer to FIG.20, machine-learning algorithms may include, without limitation, linear discriminant analysis. Machine-learning algorithm may include quadratic discriminant analysis. Machine-learning algorithms may include kernel ridge regression. Machine-learning algorithms may include support vector machines, including without limitation support vector classification-based regression processes. Machine-learning algorithms may include stochastic gradient descent algorithms, including classification and regression algorithms based on stochastic gradient descent. Machine-learning algorithms may include nearest neighbors algorithms. Machine-learning algorithms may include various forms of latent space regularization such as variational regularization. Machine-learning algorithms may include Gaussian processes such as Gaussian Process Regression. Machine-learning algorithms may include cross-decomposition algorithms, including partial least squares and/or canonical correlation analysis. Machine-learning algorithms may include naïve Bayes methods. Machine- learning algorithms may include algorithms based on decision trees, such as decision tree classification or regression algorithms. Machine-learning algorithms may include ensemble methods such as bagging meta-estimator, forest of randomized trees, AdaBoost, gradient tree boosting, and/or voting classifier methods. Machine-learning algorithms may include neural net algorithms, including convolutional neural net processes. Still referring to FIG.20, a machine-learning model and/or process may be deployed or instantiated by incorporation into a program, apparatus, system and/or module. For instance, and without limitation, a machine-learning model, neural network, and/or some or all parameters thereof may be stored and/or deployed in any memory or circuitry. Parameters such as coefficients, weights, and/or biases may be stored as circuit-based constants, such as arrays of wires and/or binary inputs and/or outputs set at logic “1” and “0” voltage levels in a logic circuit to represent a number according to any suitable encoding system including twos complement or the like or may be stored in any volatile and/or non-volatile memory. Similarly, mathematical operations and input and/or output of data to or from models, neural network layers, or the like 145 Attorney Docket No.1519-029PCT1 may be instantiated in hardware circuitry and/or in the form of instructions in firmware, machine-code such as binary operation code instructions, assembly language, or any higher- order programming language. Any technology for hardware and/or software instantiation of memory, instructions, data structures, and/or algorithms may be used to instantiate a machine- learning process and/or model, including without limitation any combination of production and/or configuration of non-reconfigurable hardware elements, circuits, and/or modules such as without limitation ASICs, production and/or configuration of reconfigurable hardware elements, circuits, and/or modules such as without limitation FPGAs, production and/or of non- reconfigurable and/or configuration non-rewritable memory elements, circuits, and/or modules such as without limitation non-rewritable ROM, production and/or configuration of reconfigurable and/or rewritable memory elements, circuits, and/or modules such as without limitation rewritable ROM or other memory technology described in this disclosure, and/or production and/or configuration of any computing device and/or component thereof as described in this disclosure. Such deployed and/or instantiated machine-learning model and/or algorithm may receive inputs from any other process, module, and/or component described in this disclosure, and produce outputs to any other process, module, and/or component described in this disclosure. Continuing to refer to FIG.20, any process of training, retraining, deployment, and/or instantiation of any machine-learning model and/or algorithm may be performed and/or repeated after an initial deployment and/or instantiation to correct, refine, and/or improve the machine- learning model and/or algorithm. Such retraining, deployment, and/or instantiation may be performed as a periodic or regular process, such as retraining, deployment, and/or instantiation at regular elapsed time periods, after some measure of volume such as a number of bytes or other measures of data processed, a number of uses or performances of processes described in this disclosure, or the like, and/or according to a software, firmware, or other update schedule. Alternatively or additionally, retraining, deployment, and/or instantiation may be event-based, and may be triggered, without limitation, by user inputs indicating sub-optimal or otherwise problematic performance and/or by automated field testing and/or auditing processes, which may compare outputs of machine-learning models and/or algorithms, and/or errors and/or error functions thereof, to any thresholds, convergence tests, or the like, and/or may compare outputs of processes described herein to similar thresholds, convergence tests or the like. Event-based 146 Attorney Docket No.1519-029PCT1 retraining, deployment, and/or instantiation may alternatively or additionally be triggered by receipt and/or generation of one or more new training examples; a number of new training examples may be compared to a preconfigured threshold, where exceeding the preconfigured threshold may trigger retraining, deployment, and/or instantiation. Still referring to FIG.20, retraining and/or additional training may be performed using any process for training described above, using any currently or previously deployed version of a machine-learning model and/or algorithm as a starting point. Training data for retraining may be collected, preconditioned, sorted, classified, sanitized or otherwise processed according to any process described in this disclosure. Training data may include, without limitation, training examples including inputs and correlated outputs used, received, and/or generated from any version of any system, module, machine-learning model or algorithm, apparatus, and/or method described in this disclosure; such examples may be modified and/or labeled according to user feedback or other processes to indicate desired results, and/or may have actual or measured results from a process being modeled and/or predicted by system, module, machine-learning model or algorithm, apparatus, and/or method as “desired” results to be compared to outputs for training processes as described above. Redeployment may be performed using any reconfiguring and/or rewriting of reconfigurable and/or rewritable circuit and/or memory elements; alternatively, redeployment may be performed by production of new hardware and/or software components, circuits, instructions, or the like, which may be added to and/or may replace existing hardware and/or software components, circuits, instructions, or the like. Further referring to FIG.20, one or more processes or algorithms described above may be performed by at least a dedicated hardware unit 2036. A “dedicated hardware unit,” for the purposes of this figure, is a hardware component, circuit, or the like, aside from a principal control circuit and/or processor performing method steps as described in this disclosure, that is specifically designated or selected to perform one or more specific tasks and/or processes described in reference to this figure, such as without limitation preconditioning and/or sanitization of training data and/or training a machine-learning algorithm and/or model. A dedicated hardware unit 2036 may include, without limitation, a hardware unit that can perform iterative or massed calculations, such as matrix-based calculations to update or tune parameters, weights, coefficients, and/or biases of machine-learning models and/or neural networks, 147 Attorney Docket No.1519-029PCT1 efficiently using pipelining, parallel processing, or the like; such a hardware unit may be optimized for such processes by, for instance, including dedicated circuitry for matrix and/or signal processing operations that includes, e.g., multiple arithmetic and/or logical circuit units such as multipliers and/or adders that can act simultaneously and/or in parallel or the like. Such dedicated hardware units 2036 may include, without limitation, graphical processing units (GPUs), dedicated signal processing modules, FPGA or other reconfigurable hardware that has been configured to instantiate parallel processing units for one or more specific tasks, or the like, A computing device, processor, apparatus, or module may be configured to instruct one or more dedicated hardware units 2036 to perform one or more operations described herein, such as evaluation of model and/or algorithm outputs, one-time or iterative updates to parameters, coefficients, weights, and/or biases, and/or any other operations such as vector and/or matrix operations as described in this disclosure. With continued reference to FIG.20, system 2000 may use user feedback to train the machine-learning models and/or classifiers described above. For example, classifier may be trained using past inputs and outputs of classifier. In some embodiments, if user feedback indicates that an output of classifier was “bad,” then that output and the corresponding input may be removed from training data used to train classifier, and/or may be replaced with a value entered by, e.g., another user that represents an ideal output given the input the classifier originally received, permitting use in retraining, and adding to training data; in either case, classifier may be retrained with modified training data as described in further detail below. In some embodiments, training data of classifier may include user feedback. With continued reference to FIG.20, in some embodiments, an accuracy score may be calculated for classifier using user feedback. For the purposes of this disclosure, “accuracy score,” is a numerical value concerning the accuracy of a machine-learning model. For example, a plurality of user feedback scores may be averaged to determine an accuracy score. In some embodiments, a cohort accuracy score may be determined for particular cohorts of persons. For example, user feedback for users belonging to a particular cohort of persons may be averaged together to determine the cohort accuracy score for that particular cohort of persons and used as described above. Accuracy score or another score as described above may indicate a degree of retraining needed for a machine-learning model such as a classifier; system 2000 may perform a larger number of retraining cycles for a higher number (or lower number, depending on a 148 Attorney Docket No.1519-029PCT1 numerical interpretation used), and/or may collect more training data for such retraining, perform more training cycles, apply a more stringent convergence test such as a test requiring a lower mean squared error, and/or indicate to a user and/or operator that additional training data is needed. Referring now to FIG.21, an exemplary embodiment of neural network 2100 is illustrated. A neural network 2100 also known as an artificial neural network, is a network of “nodes,” or data structures having one or more inputs, one or more outputs, and a function determining outputs based on inputs. Such nodes may be organized in a network, such as without limitation a convolutional neural network, including an input layer of nodes 2104, one or more intermediate layers 2108, and an output layer of nodes 2112. Connections between nodes may be created via the process of "training" the network, in which elements from a training dataset are applied to the input nodes, a suitable training algorithm (such as Levenberg-Marquardt, conjugate gradient, simulated annealing, or other algorithms) is then used to adjust the connections and weights between nodes in adjacent layers of the neural network to produce the desired values at the output nodes. This process is sometimes referred to as deep learning. Connections may run solely from input nodes toward output nodes in a “feed-forward” network, or may feed outputs of one layer back to inputs of the same or a different layer in a “recurrent network.” As a further non-limiting example, a neural network may include a convolutional neural network comprising an input layer of nodes, one or more intermediate layers, and an output layer of nodes. A “convolutional neural network,” as used in this disclosure, is a neural network in which at least one hidden layer is a convolutional layer that convolves inputs to that layer with a subset of inputs known as a “kernel,” along with one or more additional layers such as pooling layers, fully connected layers, and the like. Referring now to FIG.22, an exemplary embodiment of a node 2200 of a neural network is illustrated. A node may include, without limitation a plurality of inputs xi that may receive numerical values from inputs to a neural network containing the node and/or from other nodes. Node may perform one or more activation functions to produce its output given one or more inputs, such as without limitation computing a binary step function comparing an input to a threshold value and outputting either a logic 1 or logic 0 output or something equivalent, a linear activation function whereby an output is directly proportional to the input, and/or a non-linear activation function, wherein the output is not proportional to the input. Non-linear activation 149 Attorney Docket No.1519-029PCT1 functions may include, without limitation, a sigmoid function of the form ^^^{^} ^^^{^} ൌ ^{^} ^_ି^ష^ given ^ put x, a tanh (hyperbolic tangent) function, of the form _ି ^{ష^} in ^{^} _{^ ^^ା^ష^}, a tanh derivative function such as ^^^ ^^^ ൌ tanh^ଶ^ ^^^, a rectified linear unit function such as ^^^ ^^^ ൌ max ^0, ^^^, a “leaky” and/or “parametric” rectified linear unit function such as ^^^ ^^^ ൌ max ^ ^^ ^^, ^^^ for some a, an ^{exponential linear units function such as ^^^ ^^^ ൌ ^ ^^ ^^ ^^ ^^ ^^ ^ 0} ^_{^^ ^^௫ െ 1^ ^^ ^^ ^^ ^^ ^ 0} ^{for some value of ^^} (this function may be replaced and/or weighted by its own derivative in some embodiments), a softmax function such as ^^ ^ ^ ^{^}^ ^^_^ ൌ _{∑^ ௫^} where the inputs to an instant layer are ^^_^, a swish function such as ^^^{^} ^^^{^} ൌ ^^ ∗ ^^^, a Gaussian error linear unit function such as f(x) = ^^൫1 ^ tanh ^_^2/ ^^^ ^^ ^ ^^ ^^^{^}^^൯ for some values of a, b, and r, and/or a scaled exponential linear _{unit ^} ^{^^^ ^^௫ െ 1^ ^^ ^^ ^^ ^^ ^} ^_{^ ^^ ^^ ^^ ^^ ^ 0 . Fundamentally, there is no limit to the} nature of functions of inputs functions. As a non-limiting illustrative example, node may perform a weighted sum of inputs using weights w_i that are multiplied by respective inputs xi. Additionally or alternatively, a bias b may be added to the weighted sum of the inputs such that an offset is added to each unit in the neural network layer that is independent of the input to the layer. The weighted sum may then be input into a function φ, which may generate one or more outputs y. Weight wi applied to an input xi may indicate whether the input is “excitatory,” indicating that it has strong influence on the one or more outputs y, for instance by the corresponding weight having a large numerical value, and/or a “inhibitory,” indicating it has a weak effect influence on the one more inputs y, for instance by the corresponding weight having a small numerical value. The values of weights wi may be determined by training a neural network using training data, which may be performed using any suitable process as described above. Still referring to FIG.22, a “convolutional neural network,” as used in this disclosure, is a neural network in which at least one hidden layer is a convolutional layer that convolves inputs to that layer with a subset of inputs known as a “kernel,” along with one or more additional layers such as pooling layers, fully connected layers, and the like. CNN may include, without limitation, a deep neural network (DNN) extension, where a DNN is defined as a neural network with two or more hidden layers. 150 Attorney Docket No.1519-029PCT1 Still referring to FIG.22, in some embodiments, a convolutional neural network may learn from images. In non-limiting examples, a convolutional neural network may perform tasks such as classifying images, detecting objects depicted in an image, segmenting an image, and/or processing an image. In some embodiments, a convolutional neural network may operate such that each node in an input layer is only connected to a region of nodes in a hidden layer. In some embodiments, the regions in aggregate may create a feature map from an input layer to the hidden layer. In some embodiments, a convolutional neural network may include a layer in which the weights and biases for all nodes are the same. In some embodiments, this may allow a convolutional neural network to detect a feature, such as an edge, across different locations in an image. It is to be noted that any one or more of the aspects and embodiments described herein may be conveniently implemented using one or more machines (e.g., one or more computing devices that are utilized as a user computing device for an electronic document, one or more server devices, such as a document server, etc.) programmed according to the teachings of the present specification, as will be apparent to those of ordinary skill in the computer art. Appropriate software coding can readily be prepared by skilled programmers based on the teachings of the present disclosure, as will be apparent to those of ordinary skill in the software art. Aspects and implementations discussed above employing software and/or software modules may also include appropriate hardware for assisting in the implementation of the machine executable instructions of the software and/or software module. Such software may be a computer program product that employs a machine-readable storage medium. A machine-readable storage medium may be any medium that is capable of storing and/or encoding a sequence of instructions for execution by a machine (e.g., a computing device) and that causes the machine to perform any one of the methodologies and/or embodiments described herein. Examples of a machine-readable storage medium include, but are not limited to, a magnetic disk, an optical disc (e.g., CD, CD-R, DVD, DVD-R, etc.), a magneto- optical disk, a read-only memory “ROM” device, a random access memory “RAM” device, a magnetic card, an optical card, a solid-state memory device, an EPROM, an EEPROM, and any combinations thereof. A machine-readable medium, as used herein, is intended to include a single medium as well as a collection of physically separate media, such as, for example, a collection of compact discs or one or more hard disk drives in combination with a computer 151 Attorney Docket No.1519-029PCT1 memory. As used herein, a machine-readable storage medium does not include transitory forms of signal transmission. Such software may also include information (e.g., data) carried as a data signal on a data carrier, such as a carrier wave. For example, machine-executable information may be included as a data-carrying signal embodied in a data carrier in which the signal encodes a sequence of instruction, or portion thereof, for execution by a machine (e.g., a computing device) and any related information (e.g., data structures and data) that causes the machine to perform any one of the methodologies and/or embodiments described herein. Examples of a computing device include, but are not limited to, an electronic book reading device, a computer workstation, a terminal computer, a server computer, a handheld device (e.g., a tablet computer, a smartphone, etc.), a web appliance, a network router, a network switch, a network bridge, any machine capable of executing a sequence of instructions that specify an action to be taken by that machine, and any combinations thereof. In one example, a computing device may include and/or be included in a kiosk. FIG.23 shows a diagrammatic representation of one embodiment of a computing device in the exemplary form of a computer system 2300 within which a set of instructions for causing a control system to perform any one or more of the aspects and/or methodologies of the present disclosure may be executed. It is also contemplated that multiple computing devices may be utilized to implement a specially configured set of instructions for causing one or more of the devices to perform any one or more of the aspects and/or methodologies of the present disclosure. Computer system 2300 includes a processor 2304 and a memory 2308 that communicate with each other, and with other components, via a bus 2312. Bus 2312 may include any of several types of bus structures including, but not limited to, a memory bus, a memory controller, a peripheral bus, a local bus, and any combinations thereof, using any of a variety of bus architectures. Processor 2304 may include any suitable processor, such as without limitation a processor incorporating logical circuitry for performing arithmetic and logical operations, such as an arithmetic and logic unit (ALU), which may be regulated with a state machine and directed by operational inputs from memory and/or sensors; processor 2304 may be organized according to Von Neumann and/or Harvard architecture as a non-limiting example. Processor 2304 may include, incorporate, and/or be incorporated in, without limitation, a microcontroller, 152 Attorney Docket No.1519-029PCT1 microprocessor, digital signal processor (DSP), Field Programmable Gate Array (FPGA), Complex Programmable Logic Device (CPLD), Graphical Processing Unit (GPU), general purpose GPU, Tensor Processing Unit (TPU), analog or mixed signal processor, Trusted Platform Module (TPM), a floating point unit (FPU), and/or system on a chip (SoC). Memory 2308 may include various components (e.g., machine-readable media) including, but not limited to, a random-access memory component, a read only component, and any combinations thereof. In one example, a basic input/output system 2316 (BIOS), including basic routines that help to transfer information between elements within computer system 2300, such as during start-up, may be stored in memory 2308. Memory 2308 may also include (e.g., stored on one or more machine-readable media) instructions (e.g., software) 2320 embodying any one or more of the aspects and/or methodologies of the present disclosure. In another example, memory 2308 may further include any number of program modules including, but not limited to, an operating system, one or more application programs, other program modules, program data, and any combinations thereof. Computer system 2300 may also include a storage device 2324. Examples of a storage device (e.g., storage device 2324) include, but are not limited to, a hard disk drive, a magnetic disk drive, an optical disc drive in combination with an optical medium, a solid-state memory device, and any combinations thereof. Storage device 2324 may be connected to bus 2312 by an appropriate interface (not shown). Example interfaces include, but are not limited to, SCSI, advanced technology attachment (ATA), serial ATA, universal serial bus (USB), IEEE 1394 (FIREWIRE), and any combinations thereof. In one example, storage device 2324 (or one or more components thereof) may be removably interfaced with computer system 2300 (e.g., via an external port connector (not shown)). Particularly, storage device 2324 and an associated machine-readable medium 2328 may provide nonvolatile and/or volatile storage of machine- readable instructions, data structures, program modules, and/or other data for computer system 2300. In one example, software 2320 may reside, completely or partially, within machine- readable medium 2328. In another example, software 2320 may reside, completely or partially, within processor 2304. Computer system 2300 may also include an input device 2332. In one example, a user of computer system 2300 may enter commands and/or other information into computer system 2300 via input device 2332. Examples of an input device 2332 include, but are not limited to, an 153 Attorney Docket No.1519-029PCT1 alpha-numeric input device (e.g., a keyboard), a pointing device, a joystick, a gamepad, an audio input device (e.g., a microphone, a voice response system, etc.), a cursor control device (e.g., a mouse), a touchpad, an optical scanner, a video capture device (e.g., a still camera, a video camera), a touchscreen, and any combinations thereof. Input device 2332 may be interfaced to bus 2312 via any of a variety of interfaces (not shown) including, but not limited to, a serial interface, a parallel interface, a game port, a USB interface, a FIREWIRE interface, a direct interface to bus 2312, and any combinations thereof. Input device 2332 may include a touch screen interface that may be a part of or separate from display 2336, discussed further below. Input device 2332 may be utilized as a user selection device for selecting one or more graphical representations in a graphical interface as described above. A user may also input commands and/or other information to computer system 2300 via storage device 2324 (e.g., a removable disk drive, a flash drive, etc.) and/or network interface device 2340. A network interface device, such as network interface device 2340, may be utilized for connecting computer system 2300 to one or more of a variety of networks, such as network 2344, and one or more remote devices 2348 connected thereto. Examples of a network interface device include, but are not limited to, a network interface card (e.g., a mobile network interface card, a LAN card), a modem, and any combination thereof. Examples of a network include, but are not limited to, a wide area network (e.g., the Internet, an enterprise network), a local area network (e.g., a network associated with an office, a building, a campus or other relatively small geographic space), a telephone network, a data network associated with a telephone/voice provider (e.g., a mobile communications provider data and/or voice network), a direct connection between two computing devices, and any combinations thereof. A network, such as network 2344, may employ a wired and/or a wireless mode of communication. In general, any network topology may be used. Information (e.g., data, software 2320, etc.) may be communicated to and/or from computer system 2300 via network interface device 2340. Computer system 2300 may further include a video display adapter 2352 for communicating a displayable image to a display device, such as display device 2336. Examples of a display device include, but are not limited to, a liquid crystal display (LCD), a cathode ray tube (CRT), a plasma display, a light emitting diode (LED) display, and any combinations thereof. Display adapter 2352 and display device 2336 may be utilized in combination with processor 2304 to provide graphical representations of aspects of the present disclosure. In 154 Attorney Docket No.1519-029PCT1 addition to a display device, computer system 2300 may include one or more other peripheral output devices including, but not limited to, an audio speaker, a printer, and any combinations thereof. Such peripheral output devices may be connected to bus 2312 via a peripheral interface 2356. Examples of a peripheral interface include, but are not limited to, a serial port, a USB connection, a FIREWIRE connection, a parallel connection, and any combinations thereof. The foregoing has been a detailed description of illustrative embodiments of the invention. Various modifications and additions can be made without departing from the spirit and scope of this invention. Features of each of the various embodiments described above may be combined with features of other described embodiments as appropriate in order to provide a multiplicity of feature combinations in associated new embodiments. Furthermore, while the foregoing describes a number of separate embodiments, what has been described herein is merely illustrative of the application of the principles of the present invention. Additionally, although particular methods herein may be illustrated and/or described as being performed in a specific order, the ordering is highly variable within ordinary skill to achieve methods, systems, and software according to the present disclosure. Accordingly, this description is meant to be taken only by way of example, and not to otherwise limit the scope of this invention. Exemplary embodiments have been disclosed above and illustrated in the accompanying drawings. It will be understood by those skilled in the art that various changes, omissions and additions may be made to that which is specifically disclosed herein without departing from the spirit and scope of the present invention. 155 Attorney Docket No.1519-029PCT1

Claims

WHAT IS CLAIMED IS: 1. A system for digitizing a slide, wherein the system comprises: an optical system comprising at least an optical sensor; and a computing device configured to: perform a scan of the slide by capturing at least a first image and capturing at least a second image, wherein performing the scan comprises: identifying a first scanning parameter; using the optical system, capturing the at least a first image as a function of the first scanning parameter; determining a first quality metric as a function of the at least a first image; determining a second scanning parameter as a function of the first quality metric; and using the optical system, capturing the at least a second image as a function of the second scanning parameter.

2. The system of claim 1, wherein: the first quality metric comprises a localization quality metric; and determining the second scanning parameter comprises identifying at least a region of interest of the slide as a function of the localization quality metric.

3. The system of claim 1, wherein: the first quality metric comprises a biopsy plane estimation quality metric; and determining the second scanning parameter comprises identifying a plane within a region of interest which contains a biological specimen as a function of the biopsy plane estimation quality metric.

4. The system of claim 1, wherein: the first quality metric comprises a focus sampling quality metric; and determining the second scanning parameter comprises identifying a focal setting at which a selected point is in focus as a function of the focus sampling quality metric.

5. The system of claim 1, wherein: the first quality metric comprises a z-stack acquisition quality metric; and 156 Attorney Docket No.1519-029PCT1 determining the second scanning parameter comprises, using the optical system, capturing a z-stack at a selected point as a function of the z-stack acquisition quality metric.

6. The system of claim 1, wherein: the at least a first image comprises a macro image; and capturing the at least a second image comprises: using the optical system, capturing a first plurality of images; and combining the first plurality of images to create the at least a second image.

7. The system of claim 6, wherein performing the scan further comprises: determining a stitching quality metric as a function of the at least a second image; determining a third scanning parameter as a function of the stitching quality metric; using the optical system, capturing a second plurality of images as a function of the third scanning parameter; and combining the second plurality of images to create at least a third image.

8. The system of claim 1, wherein performing the scan further comprises capturing the at least a first image at a first location and capturing the at least a second image at a second location.

9. The system of claim 1, wherein performing the scan further comprises: determining a second quality metric as a function of the at least a first image; determining a combination quality metric as a function of the first quality metric and the second quality metric; determining a third scanning parameter as a function of the combination quality metric; and using the optical system, capturing at least a third image as a function of the third scanning parameter.

10. The system of claim 1, wherein performing the scan further comprises algorithmically removing a banding error from the at least a second image.

11. A method of digitizing a slide, wherein the method comprises: using at least a processor, performing a scan of the slide by capturing at least a first image and capturing at least a second image, wherein performing the scan comprises: identifying a first scanning parameter; 157 Attorney Docket No.1519-029PCT1 using an optical system, capturing the at least a first image as a function of the first scanning parameter; determining a first quality metric as a function of the at least a first image; determining a second scanning parameter as a function of the first quality metric; and using the optical system, capturing the at least a second image as a function of the second scanning parameter.

12. The method of claim 11, wherein: the first quality metric comprises a localization quality metric; and determining the second scanning parameter comprises identifying at least a region of interest of the slide as a function of the localization quality metric.

13. The method of claim 11, wherein: the first quality metric comprises a biopsy plane estimation quality metric; and determining the second scanning parameter comprises identifying a plane within a region of interest which contains a biological specimen as a function of the biopsy plane estimation quality metric.

14. The method of claim 11, wherein: the first quality metric comprises a focus sampling quality metric; and determining the second scanning parameter comprises identifying a focal setting at which a selected point is in focus as a function of the focus sampling quality metric.

15. The method of claim 11, wherein: the first quality metric comprises a z-stack acquisition quality metric; and determining the second scanning parameter comprises, using the optical system, capturing a z-stack at a selected point as a function of the z-stack acquisition quality metric.

16. The method of claim 11, wherein: the at least a first image comprises a macro image; and capturing the at least a second image comprises: using the optical system, capturing a first plurality of images; and combining the first plurality of images to create the at least a second image.

17. The method of claim 16, wherein performing the scan further comprises: 158 Attorney Docket No.1519-029PCT1 determining a stitching quality metric as a function of the at least a second image; determining a third scanning parameter as a function of the stitching quality metric; using the optical system, capturing a second plurality of images as a function of the third scanning parameter; and combining the second plurality of images to create at least a third image.

18. The method of claim 11, wherein performing the scan further comprises capturing the at least a first image at a first location and capturing the at least a second image at a second location.

19. The method of claim 11, wherein performing the scan further comprises: determining a second quality metric as a function of the at least a first image; determining a combination quality metric as a function of the first quality metric and the second quality metric; determining a third scanning parameter as a function of the combination quality metric; and using the optical system, capturing at least a third image as a function of the third scanning parameter.

20. The method of claim 11, wherein performing the scan further comprises algorithmically removing a banding error from the at least a second image.

21. A method comprising: evaluating an inline quality metric associated with digitizing a slide; and in response to determining that the inline quality metric is within a predetermined range, taking a remedial action.

22. A system for digitization of tissue slides based on associations among serial sections, wherein the system is comprised of: at least a computing device, wherein the computing device is comprised of: a memory, wherein the memory stores instructions; and a processor, communicatively connected to the memory, wherein the processor is configured to: retrieve a candidate tissue map associated with a candidate tissue section; retrieve a reference tissue map associated with a reference tissue section; align the candidate tissue map to the reference tissue map; 159 Attorney Docket No.1519-029PCT1 compare the aligned candidate tissue map to the reference tissue map; and generate a regenerated candidate tissue map as a function of the reference tissue map; and a scanner, configured to scan a slide and send a digitized image of the slide to the computing device.

23. The system of claim 22, wherein the memory further includes instructions configuring the processor to: identify a candidate serial section from at least one stain type, wherein the candidate serial section is associated with a case identification number and a block identification number; identify a reference serial section based on the case identification number and the block identification number; generate the reference tissue map in response to scanning the reference serial section; and generate the candidate tissue map in response to scanning the candidate serial section.

24. The system of claim 22, wherein the candidate serial section is on a first slide and the reference serial section is on a second slide.

25. The system of claim 22 wherein the candidate serial section and the reference serial section are both on a first slide.

26. The system of claim 22, wherein the system is further comprised of at least a storage device.

27. The system of claim 22, wherein the system is further comprised of a scanned slides data repository.

28. The system of claim 22, wherein the system is further comprised of a regenerated slides data repository.

29. The system of claim 22, wherein the system is further comprised of both a scanned slides data repository and a regenerated slides data repository.

30. The system of claim 22, wherein the system instantiates a machine learning module.

31. The system of claim 22, wherein the system instantiates a neural network.

32. A method for digitization of tissue slides based on associations among serial sections, wherein the method is comprised of: receiving a candidate tissue map associated with a candidate tissue section; 160 Attorney Docket No.1519-029PCT1 receiving a reference tissue map associated with a reference tissue section, the reference tissue section having a predetermined association with the candidate tissue section; aligning the candidate tissue map to the reference tissue map; comparing the aligned candidate tissue map to the reference tissue map; and generating a regenerated candidate tissue map as a function of the reference tissue map.

33. The method of claim 32, wherein the method further comprises: identifying a candidate serial section from at least one stain type, wherein the candidate serial section is associated with a case identification number and a block identification number; identifying a reference serial section based on the case identification number and the block identification number; generating the reference tissue map in response to scanning the reference serial section; and generating the candidate tissue map in response to scanning the candidate serial section.

34. The method of claim 32, wherein the candidate serial section is on a first slide and the reference serial section is on a second slide.

35. The method of claim 32, wherein the candidate serial section and the reference serial section are both on a first slide.

36. The method of claim 32, wherein the method further includes storage and retrieval of slides data from at least a storage device.

37. The method of claim 32, wherein the method further includes storage and retrieval of slides data from a scanned slides data repository.

38. The method of claim 32, wherein the method further includes storage and retrieval of slides data from a regenerated slides data repository.

39. The method of claim 32, wherein the method further includes storage and retrieval of slides data from both a scanned slides data repository and a regenerated slides data repository.

40. The method of claim 32, wherein the method instantiates a machine learning module.

41. The method of claim 32, wherein the method instantiates a neural network.

42. A slide digitization method comprising: 161 Attorney Docket No.1519-029PCT1 receiving a candidate tissue map associated with a candidate tissue section; receiving a reference tissue map associated with a reference tissue section, the reference tissue section having a predetermined association with the candidate tissue section; aligning, by a processor, the candidate tissue map to the reference tissue map; comparing, by the processor, the aligned candidate tissue map to the reference tissue map; and in response to comparing the aligned candidate tissue map to the reference tissue map, generating, by the processor, a regenerated candidate tissue map.

43. The method of claim 42, further comprising: identifying, based on at least one stain type, a candidate serial section, the candidate serial section associated with a case identification number and a block identification number; identifying, based on the case identification number and the block identification number, a reference serial section; generating, by a processor, the reference tissue map in response to scanning the reference serial section; and generating, by the processor, the candidate tissue map in response to scanning the candidate serial section.

44. The method of claim 43, wherein the candidate serial section is on a first slide and the reference serial section is on a second slide.

45. The method of claim 43, wherein the candidate serial section and the reference serial section are both on a first slide.

46. A slide digitization system comprising: a computing device, the computing device having a processor and a memory device in communication with the processor; and a scanner configured to scan a slide and send a digitized image of the slide to the computing device, wherein the processor, when executing instructions stored in the memory device, is configured to: retrieve a candidate tissue map associated with a candidate tissue section; 162 Attorney Docket No.1519-029PCT1 retrieve a reference tissue map associated with a reference tissue section, the reference tissue section having a predetermined association with the candidate tissue section; align the candidate tissue map to the reference tissue map; compare the aligned candidate tissue map to the reference tissue map; and in response to comparing the aligned candidate tissue map to the reference tissue map, generate a regenerated candidate tissue map.

47. The system of claim 46, wherein the processor is further configured to: generate a reference tissue map in response to a scanned reference serial section, the scanned reference serial section having been scanned by the scanner from a reference serial section; and generate a candidate tissue map in response to a scanned candidate serial section, the scanned candidate serial section having been scanned by the scanner from a candidate serial section;

48. The system of claim 47, wherein the candidate serial section is on a first slide and the reference serial section is on a second slide.

49. The system of claim 47, wherein the candidate serial section and the reference serial section are both on a first slide.

50. An apparatus for inline image enrichment, wherein the apparatus comprises: circuitry configured to: receive a plurality of subject data corresponding to a subject; generate a model candidate set using the plurality of subject data; instantiate at least a model of the model candidate set; digitally capture an image of the subject, wherein digitally capturing the image comprises; identifying at least a focal point using the at least a model; and capturing the image as a function of the at least a focal point; and store the captured image in a repository.

51. The apparatus of claim 50, wherein the circuitry includes a configurable hardware circuit. 163 Attorney Docket No.1519-029PCT1

52. The apparatus of claim 50, wherein generating the candidate model set further comprises classifying the plurality of subject data to at least a candidate model of a plurality of potential candidate models.

53. The apparatus of claim 50, wherein: the plurality of subject data includes at least an element of subject metadata; and generating the candidate model set further comprises generating the candidate model set using the subject metadata.

54. The apparatus of claim 50, wherein instantiating the at least a model further comprises instantiating a locally cached model.

55. The apparatus of claim 50, wherein instantiating the at least a model further comprises: receiving a remotely cached model; and instantiating the remotely cached model.

56. The apparatus of claim 50, wherein capturing the image further comprises: reviewing an initial scan; comparing the initial scan to a confidence threshold; and performing a subsequent scan based on the comparison.

57. The apparatus of claim 50, capturing the image further comprises: capturing a plurality of individual stack views of the image; saving each individual stack view; and executing a continuum diffusion process to fuse individual stacks together.

58. The apparatus of claim 57, wherein the plurality of individual stack views correspond to a plurality of distinct focal points;

59. The apparatus of claim 50 further configured to enhance at least a viewability characteristic of the image using a machine-learning process.

60. A method for inline image enrichment, wherein the method comprises: receiving, by a configured circuitry, a plurality of subject data corresponding to a subject; generating, by the configured circuitry, a model candidate set using the plurality of subject data; instantiating, by the configured circuitry, at least a model of the model candidate set; digitally capturing, by the configured circuitry, an image of the subject, wherein digitally capturing the image comprises; 164 Attorney Docket No.1519-029PCT1 identifying, by the configured circuitry, at least a focal point using the at least a model; and capturing, by the configured circuitry, the image as a function of the at least a focal point; and storing, by the configured circuitry, the captured image in a repository.

61. The method of claim 60, wherein the circuitry includes a configurable hardware circuit.

62. The method of claim 60, wherein generating the candidate model set further comprises classifying, by the configured circuitry, the plurality of subject data to at least a candidate model of a plurality of potential candidate models.

63. The method of claim 60, wherein: the plurality of subject data includes at least an element of subject metadata; and generating the candidate model set further comprises generating, by the configured circuitry, the candidate model set using the subject metadata.

64. The method of claim 60, wherein instantiating the at least a model further comprises instantiating, by the configured circuitry, a locally cached model.

65. The method of claim 60, wherein instantiating the at least a model further comprises: receiving, by the configured circuitry, a remotely cached model; and instantiating the remotely cached model.

66. The method of claim 60, wherein capturing the image further comprises: reviewing, by the configured circuitry, an initial scan; comparing, by the configured circuitry, the initial scan to a confidence threshold; and performing a subsequent scan based on the comparison.

67. The method of claim 60, capturing the image further comprises: capturing, by the configured circuitry, a plurality of individual stack views of the image; saving, by the configured circuitry, each individual stack view; and executing, by the configured circuitry, a continuum diffusion process to fuse individual stacks together.

68. The method of claim 67, wherein the plurality of individual stack views correspond to a plurality of distinct focal points.

69. The method of claim 60 further configured to enhance, by the configured circuitry, at least a viewability characteristic of the image using a machine-learning process. 165 Attorney Docket No.1519-029PCT1

70. An apparatus for visualizing digitized slides, the apparatus comprising: at least a processor; and a memory communicatively connected to the at least processor, the memory containing instructions configuring the at least processor to: retrieve a digitized slide; determine one or more visualization components of the digitized slide; generate a virtual slide corresponding to the digitized slide based on the one or more visualization components; and display a visualization of the virtual slide.

71. The apparatus of claim 70, wherein the memory contains instructions configuring the at least processor to determine, based on metadata associated with the digitized slide, that the digitized slide is a member of a set of digitized slides associated with at least one of a patient case or a tissue block.

72. The apparatus of claim 71, wherein displaying the virtual slide comprises displaying a plurality of virtual slides, including the virtual slide, corresponding to the set of digitized slides.

73. The apparatus of claim 70, wherein the one or more visualization components include at least one of a tissue section, an artifact, or an annotation.

74. The apparatus of claim 70, wherein the memory contains instructions configuring the at least processor to determine one or more user-configurable options associated with the virtual slide based on the one or more visualization components.

75. The apparatus of claim 74, wherein the one or more user-configurable options are determined by accessing a look-up table indexed by the one or more visualization components.

76. The apparatus of claim 74, wherein the visualization is displayed via a whole slide image viewer, and wherein the one or more user-configurable options are presented to a user via a user interface of the whole slide image viewer.

77. The apparatus of claim 70, wherein the memory contains instructions configuring the at least processor to receive a request to customize the visualization. 166 Attorney Docket No.1519-029PCT1

78. The apparatus of claim 70, wherein the memory contains instructions configuring the at least processor to receive a request to display a second visualization of a different virtual slide.

79. The apparatus of claim 70, wherein the memory contains instructions configuring the at least processor to: determine a recommended set of visualization components to include in the visualization; and determine a revised set of visualization components to include in the visualization based on a user selection.

80. The apparatus of claim 70, wherein the memory contains instructions configuring the at least processor to: determine that the digitized slide corresponds to an intra-serial section slide based on a presence of a plurality of serial sections in the digitized slide; classify the plurality of serial sections into a reference serial section and one or more remaining serial sections; and align the one or more remaining serial sections to the reference serial section, yielding a plurality of aligned serial sections, wherein the visualization of the virtual slide includes the plurality of aligned serial sections.

81. The apparatus of claim 80, wherein the one or more remaining serial sections are aligned with the reference serial section by computing, independently for each of the one or more remaining serial sections, one or more registration transforms relative to the reference serial section.

82. The apparatus of claim 81, wherein the one or more registration transforms are computed based on a macro image of the digitized slide, the macro image being acquired using a macro camera and having a field of view that covers each of the plurality of serial sections.

83. The apparatus of claim 82, wherein the memory contains instructions configuring the at least processor to: store the one or more registration transforms in a non-volatile storage medium; acquire a whole slide image (WSI) having a higher magnification than the macro image; 167 Attorney Docket No.1519-029PCT1 compute based on the one or more stored registration transforms, one or more corresponding high-magnification registration transforms applicable to the WSI; apply the one or more high magnification registration transforms to the plurality of serial sections within the WSI to yield a virtual WSI having a plurality of aligned serial sections, wherein displaying the visualization of the virtual slide comprises displaying a visualization of the virtual WSI.

84. The apparatus of claim 80, wherein the plurality of aligned serial sections are displayed in the same order that the corresponding plurality of serial sections appear on the digital slide.

85. The apparatus of claim 80, wherein the plurality of aligned serial sections are spatially arranged within the visualization based on a user-selected configuration.

86. The apparatus of claim 80, wherein the plurality of aligned serial sections are spatially arranged in a compact representation such that the plurality of aligned serial sections appear closer to one another in the visualization than in the digitized slide.

87. The apparatus of claim 80, wherein the one or more visualization components include at least one annotation, wherein the at least one annotation is included in the visualization based on a user-configurable filter, and wherein aligning the one or more remaining serial sections to the reference serial section includes aligning the at least one annotation to the reference serial section.

88. A method for visualizing digitized slides comprising: retrieving, by at least a processor, a digitized slide; determining, by the at least a computer processor, at least a visualization component of the digitized slide; generating, by the at least a computer processor, a virtual slide corresponding to the digitized slide based on the at least a visualization component, displaying, by the at least a computer processor and at least a display, a visualization of the virtual slide.

89. The method of claim 88, further comprising determining, by the at least a computer processor, based on metadata associated with the digitized slide, that the digitized slide is 168 Attorney Docket No.1519-029PCT1 a member of a set of digitized slides associated with at least one of a patient case or a tissue block.

90. The method of claim 89, wherein displaying the virtual slide comprises displaying a plurality of virtual slides, including the virtual slide, corresponding to the set of digitized slides.

91. The method of claim 88, wherein the at least a visualization component include at least one of a tissue section, an artifact, or an annotation.

92. The method of claim 88, further comprising determining, by the at least a computer processor, at least a user-configurable option associated with the virtual slide based on the at least a visualization component.

93. The method of claim 92, wherein the at least a user-configurable option is determined by accessing a look-up table indexed by the at least a visualization component.

94. The method of claim 92, wherein the visualization is displayed via a whole slide image viewer, and wherein the at least a user-configurable option is presented to a user via a user interface of the whole slide image viewer.

95. The method of claim 88, further comprising receiving, by the at least a computer processor, a request to customize the visualization.

96. The method of claim 88, further comprising receiving, by the at least a computer processor, a request to display a second visualization of a different virtual slide.

97. The method of claim 88, further comprising: determining, by the at least a computer processor, a recommended set of visualization components to include in the visualization; and determining, by the at least a computer processor, a revised set of visualization components to include in the visualization based on a user selection.

98. The method of claim 88, further comprising: determining, by the at least a computer processor, that the digitized slide corresponds to an intra-serial section slide based on a presence of a plurality of serial sections in the digitized slide; classifying, by the at least a computer processor, the plurality of serial sections into a reference serial section and at least a remaining serial section; and 169 Attorney Docket No.1519-029PCT1 aligning, by the at least a computer processor, the at least a remaining serial section to the reference serial section, yielding a plurality of aligned serial sections, wherein the visualization of the virtual slide includes the plurality of aligned serial sections.

99. The method of claim 98, wherein the at least a remaining serial section is aligned with the reference serial section by computing, independently for each of the at least a remaining serial section, at least a registration transform relative to the reference serial section.

100. The method of claim 99, wherein the at least a registration transform is computed based on a macro image of the digitized slide, the macro image being acquired using a macro camera and having a field of view that covers each of the plurality of serial sections.

101. The method of claim 100, further comprising: storing, by the at least a computer processor, the at least a registration transform in a non- volatile storage medium; acquiring, by the at least a computer processor, a whole slide image (WSI) having a higher magnification than the macro image; computing, by the at least a computer processor, based on the at least a stored registration transform, at least a corresponding high-magnification registration transform applicable to the WSI; applying, by the at least a computer processor, the at least a high magnification registration transform to the plurality of serial sections within the WSI to yield a virtual WSI having a plurality of aligned serial sections wherein displaying the visualization of the virtual slide comprises displaying a visualization of the virtual WSI.

102. The method of claim 98, wherein the plurality of aligned serial sections are displayed in the same order that the corresponding plurality of serial sections appear on the digital slide.

103. The method of claim 98, wherein the plurality of aligned serial sections are spatially arranged within the visualization based on a user-selected configuration.

104. The method of claim 98, wherein the plurality of aligned serial sections are spatially arranged in a compact representation such that the plurality of aligned serial sections appear closer to one another in the visualization than in the digitized slide. 170 Attorney Docket No.1519-029PCT1

105. The method of claim 98, wherein the at least a visualization component includes at least one annotation, wherein the at least one annotation is included in the visualization based on a user-configurable filter, and wherein aligning the at least a remaining serial section to the reference serial section includes aligning the at least one annotation to the reference serial section.

106. An apparatus for imaging a slide, the apparatus comprising: at least an optical system, including an optical sensor; a slide port configured to hold a slide; at least a processor; and a memory communicatively connected to the at least processor, the memory containing instructions configuring the at least processor to: receive at least a region of interest; capture, using the at least an optical system, a first image of the slide at a first position within the at least a region of interest; identify a focus pattern as a function of the first image and the first position; extrapolate a focus distance for a second position as a function of the focus pattern; and capture, using the at least an optical system, a second image of the slide at a second position and at the focus distance.

107. The apparatus of claim 106, further comprising an actuator mechanism mechanically connected to a mobile element; and wherein the memory contains instructions configuring the at least processor to, using the actuator mechanism, move the mobile element into the second position.

108. The apparatus of claim 106, wherein identifying the focus pattern comprises: identifying a row; capturing a plurality of first images at a first location, wherein each of the plurality of first images has a different focus distance; determining an optimally focused first image of the plurality of first images having an optimal focus; and identifying the focus pattern using a focus distance of a plurality of optimally focused images at a set of points along the row. 171 Attorney Docket No.1519-029PCT1

109. The apparatus of claim 108, wherein the row further includes the second position and the instructions further configure the processor to: capture, using the optical system, a plurality of second images at a second location, wherein each of the plurality of second images have a different focus distance; determine an optimally focused second image of the plurality of second images having an optimal focus; identify the focus pattern using the focus distance of the optimally focused first image and the optimally focused second image; and extrapolate a third focus distance for a third position as a function of the focus pattern.

110. The apparatus of claim 109, wherein the third position is located outside of the row.

111. The apparatus of claim 109, wherein the third position is located within a different region of interest than the first position.

112. The apparatus of claim 108, wherein identifying the row comprises identifying the row based on a first row sample presence score from a first set of row sample presence scores.

113. The apparatus of claim 112, wherein identifying the row based on the first row sample presence score comprises determining the row whose adjacent rows have the highest sample presence scores from a second set of sample presence scores, wherein the second set of sample presence scores is determined using machine vision.

114. The apparatus of claim 108, wherein the instructions further configure the processor to identify a point within the row that has a maximum point sample presence score, using machine vision.

115. The apparatus of claim 108, wherein identifying a plane comprises: identifying a plurality of points and a plurality of optimal focuses at the plurality of points; and generating the plane as a function of a subset of the plurality of points and a corresponding subset of optimal focuses at those points.

116. The apparatus of claim 115, wherein identifying the focus pattern further comprises updating the focus pattern, wherein updating the focus pattern comprises: identifying an additional point and an optimal focus at the additional point; and updating the focus pattern as a function of the additional point and optimal focus at the additional point. 172 Attorney Docket No.1519-029PCT1

117. The apparatus of claim 106, wherein capturing the second image comprises: capturing a plurality of images taken with focus distance based on the focus pattern; and constructing the second image from the plurality of images.

118. The apparatus of claim 106, wherein the memory contains instructions configuring the at least processor to: capture, using the optical system, a low magnification image of the slide, wherein the low magnification image has a magnification lower than that of the first image; identify, using machine vision, the at least a region of interest within the low magnification image; and determine if either of the first image or the second image contains a sample.

119. A method of imaging a slide, the method comprising: using at least a processor, receiving at least a region of interest; using at least a processor and at least an optical system, capturing a first image of the slide at a first position within the at least a region of interest; using at least a processor, identifying a focus pattern as a function of the first image and the first position; using at least a processor, extrapolating a focus distance for a second position as a function of the focus pattern; and using at least a processor and the at least an optical system, capturing a second image of the slide at a second position and at the focus distance.

120. The method of claim 119, further comprising, using an actuator mechanism, moving a mobile element into the second position.

121. The method of claim 119, wherein identifying the focus pattern comprises: identifying a row which includes the first position; capturing a plurality of first images at a first location, wherein each of the plurality of first images has a different focus distance; determining an optimally focused first image of the plurality of first images having an optimal focus; and identifying the focus pattern using a focus distance of a plurality of optimally focused images at a set of points along the row. 173 Attorney Docket No.1519-029PCT1

122. The method of claim 121, wherein the row further includes the second position and the method further comprises: using at least a processor and the optical system, capturing a plurality of second images at a second location, wherein each of the plurality of second images have a different focus distance; using at least a processor, determining an optimally focused second image of the plurality of second images having an optimal focus; using at least a processor, identifying the focus pattern using the focus distance of the optimally focused first image and the optimally focused second image; and using at least a processor, extrapolating a third focus distance for a third position as a function of the focus pattern.

123. The method of claim 122, wherein the third position is located outside of the row.

124. The method of claim 122, wherein the third position is located within a different region of interest than the first position.

125. The method of claim 121, wherein identifying the row comprises identifying the row based on a first row sample presence score from a first set of row sample presence scores.

126. The method of claim 125, wherein identifying the row based on the first row sample presence score comprises determining the row whose adjacent rows have the highest sample presence scores from a second set of sample presence scores, wherein the second set of sample presence scores is determined using machine vision.

127. The method of claim 121, further comprising identifying a point within the row that has a maximum point sample presence score, using machine vision.

128. The method of claim 121, wherein identifying a plane comprises: identifying a plurality of points and a plurality of optimal focuses at the plurality of points; and generating the plane as a function of a subset of the plurality of points and a corresponding subset of optimal focuses at those points.

129. The method of claim 128, wherein identifying the focus pattern further comprises updating the focus pattern, wherein updating the focus pattern comprises: identifying an additional point and an optimal focus at the additional point; and 174 Attorney Docket No.1519-029PCT1 updating the focus pattern as a function of the additional point and optimal focus at the additional point.

130. The method of claim 119, wherein capturing the second image comprises: capturing a plurality of images taken with focus distance based on the focus pattern; and constructing the second image from the plurality of images.

131. The method of claim 119, further comprising: using at least a processor and the optical system, capturing a low magnification image of the slide, wherein the low magnification image has a magnification lower than that of the first image; using at least a processor and machine vision, identifying the at least a region of interest within the low magnification image; and using at least a processor, determining if either of the first image or the second image contains a sample. 175 Attorney Docket No.1519-029PCT1