WO2014153423A2 - Procédé et système d'analyse de spécimens biologiques par imagerie spectrale - Google Patents

Procédé et système d'analyse de spécimens biologiques par imagerie spectrale Download PDF

Info

Publication number
WO2014153423A2
WO2014153423A2 PCT/US2014/031254 US2014031254W WO2014153423A2 WO 2014153423 A2 WO2014153423 A2 WO 2014153423A2 US 2014031254 W US2014031254 W US 2014031254W WO 2014153423 A2 WO2014153423 A2 WO 2014153423A2
Authority
WO
WIPO (PCT)
Prior art keywords
image
biological sample
region
interest
prediction image
Prior art date
Application number
PCT/US2014/031254
Other languages
English (en)
Other versions
WO2014153423A3 (fr
Inventor
Stanley H. Remiszewski
Clay M. Thompson
Max Diem
Aysegul EGIN
Original Assignee
Cireca Theranostics, Llc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Cireca Theranostics, Llc filed Critical Cireca Theranostics, Llc
Priority to MX2015013415A priority Critical patent/MX2015013415A/es
Priority to KR1020157030157A priority patent/KR20160012114A/ko
Priority to AU2014235921A priority patent/AU2014235921A1/en
Priority to BR112015024134A priority patent/BR112015024134A2/pt
Priority to JP2016504352A priority patent/JP2016514869A/ja
Priority to EP14768134.0A priority patent/EP2976735A4/fr
Priority to CA2907405A priority patent/CA2907405A1/fr
Publication of WO2014153423A2 publication Critical patent/WO2014153423A2/fr
Priority to IL241496A priority patent/IL241496B/en
Publication of WO2014153423A3 publication Critical patent/WO2014153423A3/fr
Priority to HK16108997.8A priority patent/HK1221051A1/zh

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/60—Type of objects
    • G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
    • G06V20/698—Matching; Classification
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons

Definitions

  • a number of diseases may be diagnosed using classical cytopathology and histopathology methods involving examination of nuclear and cellular morphology and staining patterns.
  • diagnosis occurs via examining up to 10,000 cells in a biological sample and finding about 10 to 50 cells or a small section of tissue that may be abnormal. This finding is based on subjective interpretation of visual microscopic inspection of the cells in the sample.
  • the exfoliation brush was smeared onto a microscope slide, hence the name "Pap smear.” Subsequently, the cells were stained with hematoxylin/eosin (H&E) or a "Pap stain” (which consists of H&E and several other counterstains), and were inspected visually by a cytologist or cyto-technician, using a low power microscope.
  • H&E hematoxylin/eosin
  • a "Pap stain” which consists of H&E and several other counterstains
  • tissue sections rather than exfoliated individual cells, are inspected by a pathologist using a microscope after suitable staining of the tissue.
  • the pathologist focuses on gross tissue architecture, cell morphology, nuclear morphology, nucleus- to-cytoplasm ratio, chromatin distribution, presence of mitotic figures, and others. Since these criteria are morphology-based, their interpretation always will be somewhat subjective. Immuno-histochemical and other more recent methods are often used to augment the pathologist's subjective assessment of a tissue diagnosis.
  • Spectral Histopathology SHP
  • SCP spectral cytopathology
  • a broadband infrared (IR) or other light output is transmitted to a sample (e.g., a tissue sample), using instrumentation, such as an interferometer, to create an interference pattern.
  • a sample e.g., a tissue sample
  • instrumentation such as an interferometer
  • Reflected and/or transmitted light is then detected, typically as an interference pattern.
  • a Fast Fourier Transform may then be performed on the detected pattern to obtain spectral information relating to each sample pixel.
  • the resulting information is referred to as a pixel spectrum.
  • One limitation of the FFT based related art process is that the amount of radiative energy available per unit time in each band pass may be very low, due to use of a broadband infrared spectrum emission. As a result, the data available for processing with this approach is generally inherently noise limited. Further, in order to discriminate the received data from background noise, for example, with such low energy levels available, high sensitivity instruments must be used, such as high sensitivity liquid nitrogen cooled detectors (the cooling alleviates the effects of background IR interference). Among other drawbacks, such related art systems may incur great costs, footprint, and energy usage.
  • aspects of the present invention relate to systems and methods of analysis of imaging data and assessment of imaged samples, including tissue samples, to provide a classification of a biological sample into diagnosis, prognosis, predictive, and therapeutic classes. More specifically, aspects of the present invention are directed to systems and methods for receiving biological samples and providing analysis of the biological sample data to assist in medical diagnosis.
  • aspects of the present invention include methods, devices, and systems for imaging tissue and other samples using IR spectral information from non-coherent as well as coherent sources, such as a broad-band, tunable quantum cascade laser (QCL) or optical parametric oscillators (OPSs) designed for the rapid collection of infrared microscopic data for medical diagnostics across a wide range of discrete spectral increments.
  • the infrared data may be processed by an SHP system to provide analytical data, a medical diagnosis, a prognosis, and/or predictive analysis.
  • Such methods, devices, and systems may be used to detect abnormalities in biological samples, for example, before such abnormalities may be diagnosed using related art cytopathological or histopathological methods.
  • the methods, devices, and systems may be used, for example, to conveniently allow a practitioner to obtain information regarding a biological sample, including analytical data and/or a medical diagnosis.
  • the methods, devices, and systems may also be used to train one or more machine learning methods or algorithms to provide a diagnosis, prognosis, therapeutic, sub-typing, and/or predictive classification of a biological sample.
  • the methods, devices, and systems may be used to generate one or more classification models that may be used to perform a medical diagnosis, prognosis, therapeutic, sub-typing, and/or predictive analysis of a biological sample.
  • the methods, devices and systems may be used to generate a confidence value for the predictive classifications generated.
  • the confidence value may be included in a confidence prediction image.
  • the confidence value may be included in a confidence prediction report.
  • the methods, devices, and systems may also be used to identify and assign new classes and/or sub-types of cancers.
  • the methods, devices, and systems may be used to grade the predictive classifications generated.
  • the grade may provide a degree of development of the cancer, for example.
  • the predictive classifications and grade may be used to perform a medical diagnosis and/or prognosis of a biological sample.
  • the predictive classification may also be used to associate a patient to therapy populations based on the disease state (e.g., the degree of development of the cancer).
  • the methods, devices, and systems may be used to direct harvesting material for molecular gene sequencing analysis for therapy.
  • the methods, devices and systems may be used to annotate the gene expression of a biological sample.
  • Fig. 1 illustrates an example of identifying disease states using confidence values and differentiation values to aid in the identification of classes of cancers in a biological sample in accordance with an aspect of the present invention
  • the figure shows where novel new classes or sub-types might be found (i.e., high differentiated/low confidence regions or low differentiation/high confidence regions);
  • Fig. 2 is a color Photostat of an example of SHP analysis of a region in a biological sample where two different diagnostic regions (blue and magenta) abut, and where the regions of diagnostic uncertainty (low confidence) are indicated by white areas in accordance with an aspect of the present invention;
  • FIG. 3 illustrates an example of the method flow for analyzing a spectral dataset from a biological specimen to provide a diagnosis, prognosis, and/or predictive classification of a disease or condition in accordance with an aspect of the present invention as well as identify novel new classes and subtypes;
  • FIG. 4 illustrates an example method flow for using an SHP image to locate and identify regions of a biological sample for micro-dissection in accordance with an aspect of the present invention
  • FIGs. 5A and 5B illustrate an example method flow for preprocessing IR image data in accordance with an aspect of the present invention
  • Fig. 6A is a color Photostat of an example true image (the actual annotation) in accordance with an aspect of the present invention.
  • Fig. 6B is a color Photostat of an example of a SHP prediction image in accordance with an aspect of the present invention.
  • Fig. 6C is a color Photostat of an example of a confidence prediction image in accordance with an aspect of the present invention.
  • Figs. 7A and 7B are color Photostats of example confidence prediction images in accordance with an aspect of the present invention
  • Fig. 7C is a color Photostat of an example prediction overlay on a clinical image in accordance with an aspect of the present invention
  • Fig. 8A is a color Photostat of an example of a true image, based on pathology-based annotation, in accordance with an aspect of the present invention.
  • Fig. 8B is a color Photostat of an example of a prediction image in accordance with an aspect of the present invention.
  • Fig. 8C is a color Photostat of an example image after a true positive/true negative and false positive/false positive analysis has been applied to a prediction image, in accordance with an aspect of the present invention
  • Figs. 9A and 9B are color Photostats of images with regions of interest selected (some of which are poorly differentiated) in accordance with an aspect of the present invention.
  • Fig. 10 is a color Photostat of an example confidence prediction image in accordance with an aspect of the present invention.
  • Fig. 1 1 is a color Photostat of an example true image in accordance with an aspect of the present invention.
  • Fig. 12 is a color Photostat of an example prediction image in accordance with an aspect of the present invention.
  • Fig. 13 is a color Photostat of an example legend of the confidence scale associated with a confidence prediction image (e.g., Figs. 10-13) in accordance with an aspect of the present invention
  • Fig. 14A is a color Photostat of an example classification of a biological sample in accordance with an aspect of the present invention
  • Fig. 14B is a color Photostat of an example classification of benign and malignant tumors in accordance with an aspect of the present invention
  • Fig. 14C illustrates an example algorithm structure to use in accordance with an aspect of the present invention where A, B, C and D may indicate certain tissue conditions, classes or sub-types;
  • Fig. 15 is a color Photostat of an example classification report in accordance with an aspect of the present invention.
  • Fig. 16 illustrates an example validation report in accordance with an aspect of the present invention
  • Fig. 17 is a color Photostat of example micro-dissection selection regions in accordance with an aspect of the present invention.
  • Figs. 18A- 8D are color Photostats of example of the identification and localization of micro-dissection selection regions in accordance with an aspect of the present invention.
  • Fig. 19 shows various features of a computer system for use in conjunction with aspects of the invention.
  • Fig. 20 shows an example computer system for use in conjunction with aspects of the invention.
  • aspects of the present invention include methods, systems, and devices for classifying a biological sample into diagnosis, prognosis, and therapeutic classes to provide analytical data, medical diagnosis, prognosis, therapeutic and/or predictive analysis of a biological sample.
  • the methods, devices and systems may be used to generate a confidence value for the predictive classifications generated, for example.
  • a confidence value may illustrate a level of confidence that a disease may be present in a biological sample or regions of a biological sample.
  • the confidence value may illustrate a 90% level of confidence that a disease may be present in a biological sample.
  • the confidence value may illustrate a 3% level of confidence that a disease may be present in a biological sample.
  • a confidence value may be included in a confidence prediction image.
  • the confidence prediction image may include a visual representation of a confidence value across a biological sample or within a region of a biological sample.
  • the confidence images may be used adjectively to aid a medical practitioner in providing a diagnosis.
  • the confidence images may be used to drive areas of interest for micro-dissection.
  • the confidence images and confidence values reports may also be used, for example, to visibly illustrate overlapping disease states and/or margins of the disease types for heterogeneous diseases and the level of confidence associated with the overlapping disease states.
  • a medical profession may be able to use to the prediction report to identify a prominent disease identified in a biological sample, along with any other diseases that may be present in the biological sample.
  • the methods, devices, and systems may also be used to grade the cancer identified in the predictive classifications generated (e.g., the class and/or sub-class of cancer identified).
  • the grade may provide a degree of development of the cancer from an early stage of development to a well- developed cancer.
  • the grade may be a numerical grade, such as Grade 1 , Grade 2, Grade 3, etc.
  • the grade may be described, for example, in reference to a range, such as a "High Grade,” a "Low Grade", and an "Intermediate Grade.”
  • a grade of a disease may be determined through a subjective interpretation of biological sample by a pathologist. However, the system may apply a grade to the predictive classifications generated for a biological sample.
  • the system may receive a biochemical signature of the biological sample and may use the biochemical signature to determine the grade of the predictive classifications.
  • a medical professional may be able to receive a report with one or more classes and/or sub-classes of cancers identified in a biological sample, along with a level of development for the classes and/or sub-classes of the cancers identified.
  • the predictive classifications and grade(s) may be used by the system to provide a medical diagnosis and/or prognosis of a biological sample.
  • the predictive classifications and grade(s) may be used to drive therapeutic decisions relating to the biological sample.
  • the system may determine a differentiation value of the sample to aid in determining a grade or level of development of the disease.
  • the differentiation value may be a quantitative measure for the grade or level of development of the disease.
  • a low differentiation value may indicate, for example, that a particular disease identified in the sample has not developed, a disease is in an early stage of development, and/or a different type of disease may be present.
  • a medium differentiation value may, for example, indicate that a particular disease is developing in the biological sample.
  • a high differentiation value may, for example, indicate that a particular disease present in the biological sample may be more developed.
  • a low differentiation value may indicate, for example, regions of the sample that are poorly differentiated, while a high differentiation value may indicate regions of the sample that are well differentiated.
  • a poorly differentiated region may include a region of the true image where the information identified in the biological sample (e.g., morphologic features) used to diagnose a disease may not be well developed.
  • a true image may include, for example, an annotated image by a medical professional indicating a type of disease, such as a class and/or sub-class of cancer, if any, maybe be present in a biological sample.
  • a poorly differentiated region may occur where information may not be crisp in the image.
  • Figs. 9A and 9B illustrate example images with poorly differentiated regions.
  • a highly differentiated region may include a region of the true image where the information identified in the biological sample used to diagnose disease may be well developed. For example, a highly differentiated region may occur where the information appears to be a class or sub-class of cancer.
  • the system may determine whether a region of a biological sample is poorly differentiated or highly differentiated by analyzing an annotation associated with a true image of the sample. For example, a medical professional may annotate a true image of the sample by identifying region(s) of an image where the features are a low quality, medium quality, or a high quality.
  • the system may also determine whether a region of a sample is poorly differentiated or highly differentiated through spectral analysis of the prediction image. For example, the system may determine an area of the spectra with a poor performance. In addition, the system may determine an area of the spectra with a high performance.
  • the methods, devices, and systems may be used to identify normal areas of a biological sample, classes, and/or sub-classes of cancers present in a biological sample, and/or assign new classes or sub-types of cancers to the biological sample, as illustrated in Fig. 1.
  • Fig. 1 illustrated is an example graph 100 for using confidence values and differentiation values to aid in the identification of classes of cancers in a biological sample, in accordance with an aspect of the present invention.
  • Graph 100 illustrates on the y-axis a range of differentiation values from 1 to 10, where 1 represents a poorly differentiated sample and 10 represents a well differentiated sample.
  • graph 100 illustrates on the x-axis a range of confidence values from 1 to 10, where 1 represents a low confidence value and 10 represents a high confidence value.
  • Transitional regions may include regions of the biological sample where disease(s) may be starting to develop in the biological sample. Different diseases may appear similar in early developmental stages. As such, transitional regions may identify a plurality of diseases in the biological sample. Pure regions may include regions of the biological sample where a disease is highly developed.
  • FIG. 2 illustrated therein is an example analysis of a transitional region in accordance with an aspect of the present invention.
  • Fig. 2 illustrates an example confidence prediction image (e.g., the image on the far right of Fig. 2) that may be generated by the system based on the analysis of a transitional region of the biological sample identified in Fig. 1 .
  • the confidence prediction image illustrates how confident the system may be that a particular class of cancer may be present in the transitional region.
  • new classes may be identified in a biological sample when a sample is well differentiated but there may be a low confidence level for the type of class or sub-class identified in the biological sample.
  • a new class may be identified in a biological sample when a sample is poorly differentiated, but where a high confidence level is present.
  • a high confidence level may be determined, for example, via spectral analysis. For example, a signal where the spectra from the prediction image is pure (e.g., the signal is not mixed with other spectra), the confidence level may be high. However, a signal where the spectra from the prediction image is mixed (e.g., the signal is mixed with other spectra from different classes), the confidence level may be lower.
  • New classes may also be identified in a biological sample when a disagreement occurs between a true image (e.g., an annotated image by a medical professional indicating what type of cancer, if any, is present in a biological sample) and a prediction image (e.g., a spectral image indicating what type of cancer, if any, is present in a biological sample based upon spectral analysis), as discussed in further detail below in conjunction with Fig. 3.
  • a true image e.g., an annotated image by a medical professional indicating what type of cancer, if any, is present in a biological sample
  • a prediction image e.g., a spectral image indicating what type of cancer, if any, is present in a biological sample based upon spectral analysis
  • a confidence value may be used in cooperation with the differentiation value to identify a class or classes of cancer present in a biological sample. Identifying new classes of cancer is discussed in more detail in conjunction with Fig. 3.
  • Fig. 3 illustrated therein is an example method flow 300 for analyzing a biological specimen to provide a diagnosis, prognosis, and/or predictive classification of a disease or condition, in accordance with an aspect of the present invention. Method 300 may also be used for identifying sub-classifications of cancer, in accordance with an aspect of the present invention.
  • method 300 may also be used for differentiation of normal tissue with respect to a cancerous lesion (e.g., a differentiation of normal tissue proximal to a cancerous lesion and normal tissue at a distal location from the cancerous lesion), in accordance with an aspect of the present invention.
  • a cancerous lesion e.g., a differentiation of normal tissue proximal to a cancerous lesion and normal tissue at a distal location from the cancerous lesion
  • the method may include receiving biological samples 302.
  • the biological sample may include tissue or cellular material from an individual, such as a human or animal.
  • the biological sample may be obtained by a practitioner via any known methods.
  • the sample may, for example, include a microtome section of tissue from, among other sources, biopsies, a deposit of cells from a sample of exfoliated cells, or Fine Needle Aspiration (FNA).
  • FNA Fine Needle Aspiration
  • this disclosure is not limited to these biological samples, but may include any sample for which spatially resolved infrared spectroscopic information may be desired.
  • a variety of cells or tissues may be examined using the present methodology.
  • Such cells may comprise exfoliated cells, including epithelial cells.
  • Epithelial cells are categorized as squamous epithelial cells (simple or stratified, and keratinized, or non-keratinized), columnar epithelial cells (simple, stratified, or pseudostratified; and ciliated or nonciliated), and cuboidal epithelial cells (simple or stratified, ciliated or nonciliated).
  • These epithelial cells line various organs throughout the body, such as the intestines, ovaries, male germinal tissue, the respiratory system, cornea, nose, and kidney.
  • Glandular epithelial cells are a type of epithelial cell that may be found lining the throat, stomach, blood vessels, the lymph system, and the tongue.
  • Mesothelial cells are a type of epithelial cell that may be found lining body cavities.
  • Urothelial cells are a type of epithelial cell that may be found lining the bladder. Endothelial cells are found lining blood vessels.
  • the system may have a receiving module operable to receive the biological sample.
  • the system may receive data corresponding to the biological sample.
  • an individual may provide data corresponding to the biological sample to the system.
  • the method may also include generating a spectral image of the biological sample 304.
  • the system may collect spectral data of biological sample to generate a spectral image of the biological sample.
  • Spectral data may include any suitable data that is based on methods including, but not limited to infrared, Raman and related techniques such as surface or tip enhanced Raman as well as non-linear Raman techniques such as coherent anti-Stokes Raman and stimulated femtosecond Raman effect, visible, terahertz, and fluorescence spectroscopy.
  • Infrared spectroscopy may include, but is not limited to, attenuated total reflectance Fourier transform infrared spectroscopy (ATR-FTIR) as well as other infrared reflectance measurements.
  • ATR-FTIR attenuated total reflectance Fourier transform infrared spectroscopy
  • infrared spectroscopy may be used because of its fingerprint sensitivity, which is also exhibited by Raman spectroscopy.
  • Infrared spectroscopy may be used with larger tissue sections and to provide a dataset with a more manageable size than Raman spectroscopy, for example.
  • Raman spectroscopy Raman spectroscopy
  • infrared spectroscopy data may be more amenable to fully automatic data acquisition and interpretation.
  • infrared spectroscopy has the necessary sensitivity and specificity for the detection of various tissue structures and diagnosis of disease.
  • the spectral data may be obtained by the practitioner through a tunable laser-based infrared imaging system and method, which is described in related U.S. Patent Application No. 13/084,287.
  • the data may be obtained by using an infrared tunable laser as a coherent light source, for example.
  • the wavelength of IR transmissions from the tunable laser may be varied in discrete steps across a spectrum of interest, and the transmitted and/or reflected transmissions across the spectrum may be detected and used in image analysis.
  • the data may also be obtained from a commercial Fourier transform infrared spectroscopy (FTIR) system using a non-laser based light source, such as a globar, synchrotron or other broad band light source.
  • FTIR Fourier transform infrared spectroscopy
  • One example laser usable in accordance with aspects of the present invention is a quantum cascade laser (QCL), which may allow variation in IR wavelength output between about five and 12 pm, for example.
  • QCL quantum cascade laser
  • An array detector may be used to detect transmitted and/or reflected IR wavelength image information.
  • the beam of the QCL is optically conditioned to provide illumination of a macroscopic spot (ca. 5 - 8 mm in diameter) on an infrared reflecting or transmitting slide, on which the infrared beam interacts with the sample.
  • the reflected or transmitted infrared beam is projected, via suitable image optics, to an infrared array detector, which samples the complete illuminated area at a pixel size smaller than or about the same as the diffraction limit.
  • the infrared spectra of voxels of tissue or cells represent a snapshot of the entire chemical or biochemical composition of the sample voxel.
  • This infrared spectrum is the spectral data used to generate a spectral image 304. While the above description serves as a summary example of how and what spectral data may be obtained, a more detailed disclosure of example steps involved in obtaining the data is provided in U.S. Patent Application No. 13/084,287.
  • the SHP system may have a receiving module operable to receive the transmitted data.
  • the data may be automatically or manually entered into an electronic device capable of transmitting data, such as a computer, mobile telephone, personal digital assistant (PDA), or other hand-held device, and the like.
  • the SHP system may include a computer located at a remote site having appropriate algorithms to analyze the data.
  • the SHP system may include a computer located within the same local area network as the electronic device into which the data has been entered or may be on the same electronic device into which the data has been entered (e.g., the practitioner may enter the data directly into the device that analyzes the data). If the SHP system is located remotely from the electronic device, the data may be transferred to the SHP system via any suitable electronic transferring methods, such as to a local computer via a local area network, or over the Internet.
  • An example network layout and system for communicating the data to the SHP system is described in more detail below with respect to Figs. 19 and 20.
  • the sample itself may be sent to the SHP system.
  • the SHP system may have a receiving module operable to receive the sample.
  • a practitioner operating the SHP system may instead obtain the spectral data.
  • the biological sample may be physically delivered to the SHP system, for example, at the remote site, instead of just spectral data being delivered.
  • the practitioner may still provide the clinical data, when applicable.
  • the method may further include performing preprocessing on the spectral image 306. Subjecting the spectral data to a form of preprocessing may be helpful, for example, in isolating the data pertaining to the cellular material of interest and to removing confounding spectral features, as discussed in more detail in related U.S. Patent Application No. 13/067,777.
  • the method may include loading initial IR image data 502.
  • the system may load IR image data received by the system and/or previously stored in the system.
  • the system may convert the IR image data received into absorbance units and calculate spectral parameters.
  • the method may also include selecting binned data 504 and saving the dataset 506.
  • the system may bin the image data to reduce the number of pixels. Reducing the number of pixels may, for example, enhance signal to noise or other characteristics in the data.
  • the system may load the image file using 2x2 binning options.
  • the system may store the dataset into a data store.
  • the method may include removing any offset 508 and correcting for data minimum in select range(s) 510.
  • the system may remove any baseline offset from the spectral data by processing the data from frequencies in the range.
  • the method may include creating a grayscale image by integrating over select range(s) 512.
  • the system may create the grayscale image dataset by integrating spectral intensities between certain limits).
  • a grayscale image may allow pixels with any significant infrared intensity to be viewed before any filters are applied to the image.
  • the method may also include loading water vapor correction 516.
  • the system may load water vapor correction information to apply to spectra to correct for water vapor effects in the spectral image data.
  • the method may include applying water vapor correction 514 and saving the dataset 518.
  • the system may use Multiplicative Signal Correction (MSC) to correct for residual water vib-rotational contributions.
  • MSC Multiplicative Signal Correction
  • the method may include providing noise reference 520 and saving the dataset 522.
  • the system may separate the image dataset into noise and signal regions. For example, the system may assign the black areas of the grayscale image to the noise regions, and assign the shades of grey to the signal region.
  • the separation of the noise and signal regions may be based on an integration of any spectral feature between two limiting wavelengths. When the integration value in this range exceeds a minimum value in the signal, the noise spectra may be defined if the integrated intensity is between the two specified limits.
  • the method may include applying multivariate noise reduction 524.
  • the system may perform principal component analysis (PCA) on the noise spectra that may be used to order the eigenvectors of the spectra in an order of decreasing noise contributions and reconstruct signal spectra of the dataset as the sum of the eigenvectors.
  • PCA principal component analysis
  • the method may include verifying full range signal shape and power 530.
  • the method may also include rejecting "bad" data (e.g., data falling outside the range between the minimum and maximum values) 532 and saving the dataset 534.
  • the system may perform one or more quality tests to verify the signal shape and power. Quality tests may include, but are not limited to, peak frequency location, band shape, total signal intensity, and band area ratios.
  • the system may perform a quality test based on peak frequency.
  • the system may perform a quality test based on total spectral integrated intensity by integrating between certain limits. The integrated intensity may be required to fall between a certain minimum value and a certain maximum value.
  • the system may perform quality tests based on spectral integrated intensity within certain limits. Integrated intensities may be required to fall between a certain minimum value and a certain maximum value. In an aspect, the system may perform quality tests based on a peak area ratio between amide I and amide II bands by integrating between certain limits. In an example implementation, intensity ratios may be required to be between a certain minimum value and a certain maximum value.
  • the method may also include reporting regional signal to noise 536 and saving the signal to noise values 538.
  • the system may receive inputs to calculate the signal to noise of the data.
  • the inputs may include, for example, the left and right margins of a baseline region of interest, the left and right margins of the signal region of interest, and the left and right margins of the noise region of interest.
  • the method may further include enhancing signal for class separation 560 and saving the dataset 562.
  • the system may apply a smoothing derivative to smooth, for example, window width, order, and derivative.
  • the system may also apply a normalization to enhance the signal for class separation.
  • the method may include selecting region of interest for phase correction and saving the dataset 566.
  • the system may expand a spectral region of interest between certain values and select the spectral region of interest for phase correction.
  • the method may include performing phase correction on the selected region of interest 568.
  • Phase correction may include, for example, transforming the 512 data point 2 nd derivative spectral vector by a finite Hilbert transform (truncated FFT) and split into real and imaginary parts.
  • the system may perform a coordinate transformation and a new spectral vector may be created.
  • the system may select a phase corrected trial spectra with the highest frequency reference peak between a range of values as the corrected spectra. Phase correction is described in more detail in U.S. Patent Application No. 13/067,777.
  • the method may also include selecting a region of interest for a classifier 574.
  • the system may expand the region of interest between a range of values.
  • the method may also include applying normalization 574.
  • Normalization may include, but is not limited to, vector normalization, standard normal variate, and multiple regions.
  • the system may apply a normalization to the region of interest.
  • the method may include clustering 576.
  • the system may also perform a clustering.
  • the method may create cluster images 578 and saving the dataset 580.
  • the system may create and store cluster images.
  • the method may include different metrics 582 to create cluster images. These include correlating and distance calculations.
  • the method may include generating a validation report 589.
  • the system may create clustering metrics and correlate the cluster image to known references.
  • Clustering metrics may include, but are not limited to, k-means clustering and hierarchical cluster analysis (HCA).
  • HCA hierarchical cluster analysis
  • the system may generate a validation report.
  • An example validation report is illustrated in Fig. 16.
  • the method may further include receiving clinical information during the annotation process 310.
  • the system may receive clinical information from a medical practitioner, an electronic medical record of a patient, or other data source, such as a data repository that may include clinical data.
  • Clinical information may include, for example, any information that may be relevant to a diagnosis and/or prognoses, including the type of cells likely present in the sample, the part of the body from which the sample was taken, and the type of disease or condition likely present, among other diagnoses.
  • clinical information may include a clinical "gold standard” for accepted practices for the current state-of-the-art.
  • clinical "gold standards” may include using stains on biological samples such as, but not limited to, immuno-histochemical (IHC) stains and panels, hematoxylin stains, eosin stains, and Papanicolaou stains.
  • clinical "gold standards” may also include using a microscope to measure and identify features in a biological sample including staining patterns.
  • the method may also include receiving annotation information for the IR image 308.
  • Annotation information may include, but is not limited to, any suitable clinical data regarding the selected annotation region, such as data that may be relevant to a diagnosis, including, for example, biochemical signatures as correlated to a feature of a type of cells and/or tissues that are likely present in the sample; staining grades of the sample; intensities; molecular marker status (e.g., molecular marker status of IHC stains); the part of the body from which the sample was taken; and/or the type of disease or condition likely present.
  • the annotation information may relate to any measurable aspects on the visual image of the sample.
  • the annotation information may also include, for example, a time stamp (e.g., a date and/or time when the annotation was created), parent file annotation identifier information (e.g., whether the annotation is part of an annotation set), user information (e.g., name of user who created the annotation), cluster information, cluster spectra pixel information, cluster level information, and number of pixels in the selected region, among other information relating to the annotation.
  • a time stamp e.g., a date and/or time when the annotation was created
  • parent file annotation identifier information e.g., whether the annotation is part of an annotation set
  • user information e.g., name of user who created the annotation
  • cluster information e.g., cluster spectra pixel information
  • cluster level information e.g.
  • the user may select an annotation region of the registered spectral image and may provide the annotation information for the selected region.
  • the user may use the system to select a region of the registered image that corresponds to a biochemical signature of a disease and/or condition.
  • the user may place a boundary around an area in the spectral image where the spectra of pixels of the spectral image appear to be generally uniform (e.g., the color in the area of the spectral image is mostly the same color).
  • the boundary may identify a plurality of pixels in the spectral image that correspond to a biochemical signature of a disease or condition.
  • the user may select an annotation region based upon one or more attributes or features of the visual image.
  • the annotation region may correspond to a variety of visual attributes of the biological sample, as well as biochemical states of the biological sample. Annotation regions are discussed in more detail in U.S. Patent Application No. 13/507,386. It should also be noted that the user may select an annotation region of the registered spectral image that does not correspond to a biochemical signature of a disease or condition.
  • the system may automatically or otherwise (e.g., with some user assistance or input parameters) provide the annotation information for the selected annotation region, as discussed in more detail in U.S. Patent Application No. 13/645,970.
  • the system may provide the date and time the annotation was created, along with the cluster information for the selected region.
  • the system may automatically or otherwise select the annotation region of the registered spectral image and provide the clinical data (e.g., data that may be relevant to a diagnosis and/or prognosis, and classifications of a disease or condition) for the selected annotation region.
  • the system may review some or all of the cluster levels of the spectral image and may identify a cluster level where the spectral clusters of pixels are relatively uniform (e.g., a homogeneous spectral cluster of pixels with similar spectra, per a predetermined parameter).
  • the system may present each homogeneous spectral cluster as a single color (e.g., blue for one cluster and red for a different cluster).
  • the system may compare the identified cluster level with the cluster level for the selected annotation region of the spectral image, and, if the system determines that a match occurs, the system may determine that another level or cluster level should not be selected for the annotation region.
  • the method may include generating a true image 312.
  • a true image may be, for example, a visual image of the biological sample that may include an annotation region.
  • the visual image of the sample may be obtained using a standard visual microscope, such as of a type commonly used in pathology laboratories.
  • the microscope may be coupled to a high resolution digital camera that captures the field of view of the microscope digitally.
  • This digital real-time image may be based on the standard microscopic view of a sample, and may be indicative of tissue architecture, cell morphology, and staining patterns.
  • the image may be stained, e.g., with hematoxylin and eosin (H&E) and/or other constituents, immuno- histochemicals, Insitu-hybridization (ISH), etc., or unstained.
  • H&E hematoxylin and eosin
  • ISH Insitu-hybridization
  • Figs. 6A, 8A, and 1 Examples of true images are illustrated in Figs. 6A, 8A, and 1 1.
  • Figs. 6A and 11 illustrate an image with Adenocarcinoma (ADC) cancer regions annotated in a biological sample.
  • ADC Adenocarcinoma
  • the dark blue region of the image illustrates annotation regions in the biological sample, where a medical practitioner or other user has identified ADC in the biological sample.
  • Fig. 8A illustrates a true image of an entire biological sample with regions of ADC identified in the biological sample (e.g., the blue regions of the image).
  • the method may also include creating a classification model and training a classifier algorithm 314.
  • the system may be used to train algorithms to provide a diagnosis, prognosis and/or predictive classification of a disease or condition, such as is described in a more detailed example in U.S. Patent Application No. 13/645,970.
  • the system may be used to form one or more classification models for diagnosing diseases, as described in more detail in U.S. Patent Application No. 13/645,970.
  • a data repository may include a set of listed tissue or cellular classes. Classes may be derived from and may be listed, for example, to reflect expert opinions, group decisions, and/or individual and institutional standards.
  • the algorithms used to provide a diagnosis and/or a prognosis or predictive analysis for a biological sample may be trained to implement expert practices and standards, which may vary from institution to institution and among individuals.
  • the system may receive a query with one or more parameters for training, and testing features that may be correlated to a biological signature representative of the particular disease, condition, feature state, and/or class.
  • the parameters may include, but are not limited to, a disease or condition type (e.g., lung cancer or kidney cancer), cell or tissue class, tissue type, disease state, classification level, spectral class, and tissue location, among other parameters.
  • the system may receive the query and the parameters from a user of the system.
  • the system may automatically or otherwise determine the parameters that should be used for the focused on a particular disease or condition.
  • the training and testing features may be customized based upon the parameters received.
  • the system may extract pixels from the visual and spectral images stored in a data repository that correspond to the parameters for the training testing features. For example, the system may access the annotated images stored in the data repository, along with any suitable annotation information and/or meta-data corresponding to the annotated images. The system may compare the parameters of the query with the annotation information and/or meta-data of the annotated images. Upon a match occurring between the parameters and the annotation information and/or the meta-data, for example, the system may extract the pixels of the visual and spectral images associated with the parameters and form a training set of data. The pixels extracted for the training data may include pixels from different cells or tissues classes and/or tissue types.
  • the pixels extracted from different tissue types may be stored as part of different testing features.
  • the training data may include spectral data that is associated with specific diseases and/or conditions, and/or, for example, cell or tissue types (collectively, a "class").
  • the system may extract pixels of the visual and spectral images that may provide a meaningful representation of the disease or condition based upon the parameters provided for the training features, in order to provide a diagnosis, a prognosis, and/or predictive analysis of the disease or condition.
  • Verification tests may include, but are not limited to, quality tests and feature selection tests on the training set of data.
  • the system may utilize the methodology (e.g., algorithm) created by the training set of data in conjunction with a testing set of data to verify the accuracy of the methodology or algorithm.
  • the testing set of data may include biological samples that contain the particular disease or condition, along with biological samples that do not contain the particular disease or condition.
  • the system may verify the accuracy of the algorithm, for example, by determining whether the algorithm correctly identifies biological samples that contain the particular disease or condition and biological samples that do not contain the particular disease or condition.
  • the system may determine that the accuracy of the algorithm is high. However, when the algorithm is not able to correctly identify which biological samples from the testing data contain the disease or condition or incorrectly identifies biological samples as containing the disease or condition, the system may determine that the accuracy of the algorithm is low.
  • the results of the algorithm may be compared against an index value that may indicate the probability of whether the algorithm correctly identifies the biological samples. Index values above a threshold level may indicate a high probability that the algorithm correctly identified the biological samples, while index values below a threshold level may indicate a low probability that the algorithm correctly identifies the biological samples.
  • the system may refine the training set of data.
  • the system may increase and/or decrease the number of pixels, for example, in order to increase the likelihood of statistically relevant performance of the algorithm. It should be noted that the number of pixels that are required for the training set of data may vary based upon the type of disease or condition the algorithm is trying to diagnose and/or the cell or tissue class selected, for example.
  • the system may generate one or more trained algorithms to provide a diagnosis, a prognosis, and/or predictive analysis for the particular disease, based upon the testing features.
  • a plurality of algorithms may be generated to provide such diagnosis, a prognosis, and/or predictive analysis for a disease, based upon the received parameters.
  • multiple algorithms may be trained to diagnose lung cancer, with each algorithm trained to diagnose a particular type of lung cancer, based upon different parameters that may be correlated and coupled to a biochemical signature representative of the disease or feature state, and class of the disease.
  • the system may store the one or more trained algorithms in a data repository that also contains the annotated spectral and visual images, annotation information and/or meta-data.
  • the system may also be used to form one or more classification models for diagnosing diseases, such as is described in more detailed examples in U.S. Patent Application No. 13/645,970.
  • the system may combine various algorithms for diagnosing different forms of cancer (e.g., lung cancer, breast cancer, kidney cancer) to form one model for diagnosing cancer.
  • the classification models may also include sub-models.
  • the classification model for diagnosing cancer may have sub-models for diagnosing various forms of cancer (e.g., lung cancer, breast cancer, kidney cancer).
  • the sub-models may further include sub-models.
  • the model for diagnosing lung cancer may have multiple sub-models for identifying the type of lung cancer that may be present in the biological sample.
  • the system may perform sub-typing of lung cancer by identifying main cancer types and benign, such as Benign, Small Cell Lung Cancer (SCLC), Adenocarcinoma (ADC), Squamous Carcinoma (SQCC) and Large Cell Lung Cancer (LCLC).
  • SCLC Small Cell Lung Cancer
  • ADC Adenocarcinoma
  • SQCC Squamous Carcinoma
  • LCLC Large Cell Lung Cancer
  • the system may further identify sub-types of the main types of cancer identified and sub-types of the sub-types.
  • Sub-types may include, but are not limited to, Lepidic, Acinar, Papillary, Micropapillary, Solid.
  • the system may create one or more classification models for diagnosing diseases using the subtypes and types identified. For example, the system may classify the subtypes and types as classes of cancer in the classification models.
  • the classes of cancer may be used in diagnosing a biological sample.
  • the classes of cancer may be associated with therapy populations.
  • Therapy populations may include, for example, appropriate therapies for a disease state.
  • the classes may be associated with a patient population that responds to a particular therapy for a disease state.
  • the system may use the classification models to provide recommendations for appropriate therapies (e.g. as a companion diagnostic modality, and in conjunction with literature data mining) to treat the disease identified in the class or sub-class.
  • the system may distinguish the disease types and subtypes from normal tissue (e.g., tissue presumed to have no relevant disease.
  • the system may use the classes, for example, to distinguish heterogeneity of the biological sample.
  • the system may differentiate normal tissue proximal to a cancerous lesion and normal tissue at a distal location from the cancerous lesion, as illustrated, for example, in Fig. 14A.
  • Fig. 14 illustrates an example cancerous tissue (CA) with proximal normal (PN) tissue proximal to the cancerous tissue.
  • Fig. 14A also illustrates distal normal (DN) tissue at a distal location from the cancerous tissue and benign normal (BN) tissue located outside of the cancer lesion.
  • CA cancerous tissue
  • PN proximal normal
  • DN distal normal
  • BN benign normal
  • the system may analyze the proximal normal tissue, distal normal tissue and benign normal tissue.
  • Normal tissue within a tumor may have a different signature than benign lesions.
  • proximal normal tissue may have a different signature than distal normal tissue.
  • the signature of the proximal normal tissue may indicate emerging cancer in the proximal normal tissue
  • the signature of the distal normal tissue may indicate a different disease state in the distal normal tissue.
  • the system may use the proximity of the tissue to the cancerous tissue to measure, for example, a relevant strength of a disease, growth of a disease, and patterns of a disease.
  • the system may analyze the appropriate cell and tissue morphologic descriptor, such as stroma, connective tissue, and blood vessel wall.
  • the system may also identify variants of the types and sub-types. Variants may include modifiers that may occur along with any of the cancer types and histological subtypes, such as mucinous adenocarcinoma, colloidal, fetal (low and high grade) and enteric. In an aspect, the system may classify the variants as classes in the classification models.
  • Fig. 14B illustrates an example classification of benign and malignant tumors in accordance with an aspect of the present invention.
  • Fig. 14B illustrates an example sub-classification of Lung/Pulmonary benign lesions of Hamartoma, Sarcoidosis (granuloma) and Organizing Pneumonia types (blue) cluster separation versus lung cancer tumor normal types (red).
  • 1402 illustrates an example sub-classification of Benign separated by SHP.
  • 1404 illustrates an example sub-classification of necrosis, keratin pearls, and lepidic.
  • 1406 illustrates an example Squamous grades classified automatically by SHP.
  • 1408 illustrates an example sub-classification of adenocarcinoma.
  • the system may establish a rule set for determining an order for applying the methodologies (e.g., algorithms) within the classification model.
  • the system may establish a rule set for placing constraints on when algorithms may be used.
  • the rule set may vary based upon the diseases and/or the number of algorithms combined together to form the models, for example.
  • the system may generate one or more models for diagnosing the particular disease. It should be noted that, in addition to the above method, a variety of other methods may be used for creating a classification model for a particular disease or condition.
  • One example rule set for applying the algorithms within the classification model may include a variation reduction order, determined using hierarchical cluster analysis (HCA) or other clustering / segmentation methods.
  • HCA hierarchical cluster analysis
  • An example of HCA is described in detail in U.S. Patent Application
  • HCA identifies cellular and tissue classes that group together due to various similarities. Based on the HCA, the most effective order of the iterations, or variation reduction order, may be determined. That is, the iteration hierarchy/variation reduction order may be established based on the least to greatest variation in data, which is provided by HCA. By using HCA, based on the similarity or variance in the data, it may be determined which class of tissue or cell should be labeled and not included in the subsequent data subset, in order, for example, to remove variance and improve the accuracy of the identification.
  • Fig. 14C illustrates an example rule set for determining a classification of lung cancer, where A, B, C and D may indicate certain tissue conditions, classes or sub-types, in accordance with an aspect of the present invention.
  • the method described above may be applied. That is, the iterative process may be repeated, as illustrated, until the desired result is reached.
  • the practitioner may choose to test a sample generally for cancerous cells or for a particular type of cancer.
  • the conditions that are tested may be based on clinical data (e.g., what condition is most likely present) or by "blindly" testing against various conditions.
  • the method disclosed herein increases the accuracy of the diagnosis, and in particular, increases the accuracy even when there is little or no information regarding which conditions are likely present.
  • the method disclosed herein may be used for prognosis and/or predictive classifications of a disease or condition.
  • the method may further include generating prediction images 316.
  • the system may apply the one or more classification models and/or one or more classifier algorithms trained using the classification models to a true image, and generate a prediction image.
  • the system may apply the one or more classification models and/or classifier algorithms to a biological sample.
  • Example prediction images are illustrated in Figs. 6B and 12.
  • Fig. 6B illustrates an example where Squamous (SqCC) cancer is predicted in the magenta regions of the biological sample, and where ADC cancer is predicted in the blue regions.
  • the biological sample illustrated in Fig. 6B may be predicted to include both ADC and SqCC.
  • Fig. 12 illustrates another example prediction image.
  • Fig. 12 illustrates an image of an entire sample, with five classes of tissue predicted in the image.
  • the image illustrates SqCC in the blue regions, ADC in the magenta regions, Necrosis in the green regions, SCLC in the yellow regions, and Normal tissue in the red regions.
  • the method may include generating confidence prediction images 326.
  • Confidence prediction images may include a confidence value illustrating a level of confidence that a particular class or sub-class of cancer may be present in the prediction image. For example, a higher confidence value may indicate that one or more diseases are present in the prediction image. A higher confidence value may also indicate that a particular disease is more developed.
  • the system may analyze the spectra from the prediction image and when the spectra signal is close to a center of a class of cancer, the confidence level may be high.
  • a signal where the spectra from the prediction image is pure e.g., the signal is not mixed with other spectra
  • the confidence level may be high.
  • a lower confidence value may indicate, for example, that one or more diseases may be present in the prediction image.
  • the system may analyze the spectra signal and may determine how close the signal may be to a center of a class of cancer. For example, signals that may be within a class of cancer, but farther away from a center of the class (e.g., may be on a boundary or fringe of the spectra for a class), may overlap with another class of cancer. As such, a confidence level that a particular class of cancer may be present in a biological sample may be low.
  • signals that are farther away from a center of the class may indicate that the sample contains a new class of cancer, a different type of cancer, or a different sub-type of cancer.
  • a lower confidence value may also indicate that the disease has not developed and/or may be a different type of disease.
  • the confidence value may be a number, for example, in a range from 1 to 10, where 1 is a low to no confidence and 10 is a high confidence. In another example, the confidence value may be a number between 0 and 1 , where 0 is no confidence and 1 is high confidence.
  • the system may use one or more prediction calculations to calculate the confidence value. Prediction calculations may include, but are not limited to, Piatt Separation Plane, Random Forest, Baysian A-Priori Estimates, Artificial Neural Networks and LDA. It should be noted that a variety of prediction calculations may be used to calculate the confidence value.
  • the system may overlay a confidence value for each class or sub-class illustrated in the prediction image, and may generate a confidence prediction image illustrating the confidence value.
  • the confidence value may be represented in a binary manner, e.g., a white dot may be added to the image to represent a low confidence value, and no additional information may be added to the image with a high confidence value.
  • Another example may include a color scale to illustrate the confidence level. A lighter shade of a color or white may represent a low confidence value, while a solid color may represent a high confidence value.
  • Example confidence prediction images are illustrated in Figs. 7A-7C and 10. As illustrated in Figs.
  • white dots in the image may represent a low confidence that a particular class or sub-class of cancer may be present in the biological sample.
  • the spectra may indicate an abnormality for that region of the biological sample, but the system may have a low confidence in identifying the abnormality.
  • Fig. 7B illustrates an area where high concentrations of white pixels are grouped within a region of the biological sample.
  • the region of white pixels may represent an area where a new class or sub-class of cancer may be present in the biological sample.
  • the regions in Figs. 7A, 7B, 7D and 10 where a color is present may represent a high confidence that a particular class or sub-class is present in the biological sample.
  • Fig. 7C illustrates an example confidence image overlaid on a clinical image.
  • the system may overlay the confidence image on the clinical image, so that an individual may view the confidence image concurrently with the clinical image.
  • the system may transmit the confidence images for presentation on, for example, a field of view of a microscope (e.g., at a pathologist microscope), a display of a computing device, and/or a document or report.
  • the system may project a virtual image of the confidence prediction image into a field of view of a microscope so that an individual may view the confidence prediction image concurrently with the biological sample.
  • the system may overlay the virtual image of the confidence prediction image over the image of biological sample viewable in the microscope so that the individual may be able to receive a biochemical analysis of the biological sample from the confidence prediction image in the foreground of the field of view while viewing the image of the biological sample.
  • the confidence prediction image may highlight areas of the biological sample where the individual may want to analyze further. For example, the individual may be able to zoom in to view different regions of the biological sample based on region of interests identified in the confidence prediction images.
  • the system may allow an individual to turn on/off the confidence prediction image. By overlaying the confidence prediction over the clinical image, the system may allow an individual to leverage the power of biochemical analysis to identify predictive classifications in the biological sample when viewing the biological sample.
  • Fig. 13 illustrated therein is an example prediction legend with a confidence scale to use when viewing confidence prediction images, in accordance with an aspect of the present invention.
  • the prediction legend may illustrate various classes of cancer by color and apply a level of confidence to the color.
  • a low level of confidence may be a light shade of a color or white, while a high level of confidence may a more saturated shade of a color.
  • the prediction legend may include various classes of cancer that may be illustrated in the confidence prediction image.
  • the prediction legend may include ADC, SqCC, Necrosis, SCLC, and normal tissue, for example.
  • the prediction legend may assign a color value to each class of cancer represented in the prediction legend. For example, blue may be assigned to SqCC, magenta may be assigned to ADC, green may be assigned to Necrosis, yellow may be assigned to SCLC, and red may be assigned to normal tissue. Any number of classes of cancers and/or different types of diseases may be included in the prediction legend.
  • the various classes of cancers may be differentiated from each other in a variety of manners, color being one example.
  • the prediction legend may also include a confidence scale illustrating a confidence level for the prediction.
  • the confidence scale may range from 0 to 1 , with 0 representing little to no confidence and 1 representing high confidence.
  • the prediction legend may alter the color of the class of cancer based on the confidence level. For example, white or a light color may illustrate a low confidence level, and a darker or more saturated color may illustrate a high confidence level. As such, a light blue color may illustrate a low confidence that a biological sample may contain ADC cancer. While a dark green color may indicate a high confidence that a biological sample may contain Necrosis cancer.
  • the method may also include generating a prediction report with confidence values 328.
  • the prediction report may identify the classes and sub-classes of cancer identified in the biological sample and may provide a confidence value illustrating a level of confidence that a particular class or sub-class of cancer may be present in the prediction image.
  • the prediction report may include, for example, true images, prediction images and confidence images, as illustrated in Figs. 6A-6C, 7A-
  • Fig. 12 illustrates an example prediction image illustrating five classes of cancer identified in the biological sample.
  • Fig. 10 illustrates an example confidence image illustrating a level of confidence of the five classes of cancer identified in the biological sample.
  • the prediction report may include, for example, charts and/or graphs depicting diseases identified in the biological samples, and a level of confidence, as illustrated in Figs. 15 and 16.
  • Fig. 15 illustrates an example prediction report describing the types of tissue found in the biological sample, a predominant disease class identified, areas of tissue where the disease is identified, and a level of confidence for the analysis.
  • Fig. 15 illustrates an example bar graph that may illustrate the prediction results.
  • Fig. 16 illustrates an example validation report, in accordance with an aspect of the present invention.
  • an individual may review a prediction report to easily review the classes of cancer identified in a biological sample and a level of confidence associated with the class of cancer.
  • the confidence images and confidence values reports may also be used to visibly illustrate overlapping disease states and/or margins of the disease types for heterogenous diseases, and the level of confidence associated with the overlapping disease states.
  • a medical profession may be able to use to the prediction report to identify a prominent disease identified in a biological sample, along with any other diseases that may be present in the biological sample, for example.
  • the method may further include performing a difference analysis between the true image and the prediction image 318.
  • the system may compare the true image of a biological sample with the prediction image of the same biological sample and determine any differences that may be present between the true image and the prediction image.
  • Difference analysis may include, but is not limited to, comparing textures in the true image and prediction image, comparing the true image and the prediction image, comparing spectral variations (e.g., how much the spectra is changing, wide variation), identifying spatial locality differences (e.g., the areas of difference may be clustered together to make a larger region of a same color in an area, the areas of difference may be spread out in another color), IHC markers (e.g., + or -), molecular markers (e.g., + or -), histopathology, and any other suitable meta data or clinical data (e.g., patient information).
  • spectral variations e.g., how much the spectra is changing, wide variation
  • identifying spatial locality differences e.g., the areas of difference may be clustered together to make a larger region of a same color in an area, the areas of difference may be spread out in another color
  • IHC markers e.g., + or -
  • molecular markers e.
  • the system may apply one or more of the above mentioned difference analyses to the prediction image to identify regions of the prediction image that are different from the true image, without an explanation for the difference.
  • the system may apply one or more of the above mentioned difference analyses to the prediction image to identify regions of the prediction image that are different from the true image, without an explanation for the difference.
  • the system may compare the true image illustrated in Fig. 6A with the prediction image illustrated in Fig. 6B and determine whether any differences are illustrated.
  • the true image in Fig. 6A illustrates the biological sample as containing squamous cancer (e.g., the biological sample is the color blue).
  • the prediction image illustrates the biological sample as containing both squamous cancer (the blue color sample) and adenocarcinoma (the magenta color sample).
  • the system may determine that the magenta regions of the prediction image in Fig. 6B are different from the same regions in the true image.
  • the method may include assigning region of interest pixels to a new class 320.
  • the system may create an annotation region for the region of interest pixels, and assign the annotation region a new class based upon the difference analysis. For example, the system may determine that the magenta regions of the prediction image in Fig. 6B are different from the same regions in the true image in Fig. 6A, and may create annotation regions around the magenta regions of the prediction image to assign a new class.
  • the method may proceed to annotation (308) where a medical professional may provide an annotation to the image, for example, indicating whether the biological sample contains the new class.
  • the method may include determining a true positive region of interest or true negative region of interest 322.
  • the system may identify pixels of the comparison image that include a true positive region of interest or a true negative region of interest.
  • a true positive region may include, for example, a region of the comparison image where a true image indicates that a class of cancer is present in the true image (e.g., a medical professional annotated the true image with the class of cancer), and where the spectra from the prediction image indicate that a class of cancer is present in the prediction image.
  • a true negative may include, for example, a region of the comparison image where a true image of the biological sample indicates that a class of cancer is not present in the true image (e.g., a medical professional annotated the true image to indicate a class of cancer is not present in the true image), and where the spectra from the prediction image indicates that a class of cancer is not present in the prediction image.
  • a prediction image with true a positive region is illustrated in Fig. 8B.
  • Fig. 8A illustrates a true image of a biological sample with SqCC+ identified in the blue regions.
  • Fig. 8B illustrates a prediction image for the same biological sample identified in Fig. 8A, with true positive regions where the prediction image also identifies SqCC+ in the same regions identified in the true image.
  • the blue regions in the prediction image may correspond to the blue regions in the true image.
  • the method may also include determining any false positive region of interest and any false negative region of interest 324.
  • the system may identify pixels of the comparison image that include a false positive region of interest or a false negative region of interest.
  • a false positive region of interest may include, for example, a region in the comparison image where the true image indicates that a class of cancer is not present in the true image and the spectra from the prediction image indicates that the class of cancer is present in the prediction image.
  • a false negative region of interest may include, for example, a region in the comparison image where the true image indicates that a class of cancer is present in the true image and the spectra from the prediction image indicates that the class of cancer is not present in the prediction image.
  • FIG. 8C An example of a prediction image with a false negative region is illustrated in Fig. 8C.
  • Fig. 8C may illustrate a prediction image for the same biological sample illustrated in Fig. 8A.
  • the false positive regions illustrated in Fig. 8C may include the green regions indicating that
  • Necrosis may be present in the biological samples, where the true image, Fig.
  • the method may further include selecting a region of interest in the confidence prediction image based on confidence values 330.
  • a region of interest may include regions in the sample that are well differentiated, but where there may be a low confidence level for the type of class or sub-class identified in the biological sample.
  • Fig. 7B illustrates an example of a region of interest that is well differentiated, with a low confidence level.
  • the region of interest shown in Fig. 7B has several multiple white dots spatially located in the region of interest.
  • Figs. 9A and 9B illustrate example images with poorly differentiated regions of interest selected.
  • a region of interest may also include regions in the sample that are poorly differentiated, but where there may be a high confidence level in the spectra signal.
  • a region of interest may include a plurality of colored pixels indicating a strong spectra signal for a different class from the true image, located in a poorly differentiated area of the image.
  • the system may receive identified false negative regions of interest and false positive regions of interest and may identify the region of interest by inserting a boundary around the region of interest, such as a circle, a grid, an outline, or other forms of boundaries.
  • the method may also include assigning region of interest pixels to a new class 332.
  • the system may create an annotation region for the region of interest pixels and assign the annotation region a new class or sub-class.
  • the method may proceed to annotation (308), where a medical professional may provide an annotation to the image indicating that the biological sample may contain the new class, or other methodology (e.g., algorithm) may be applied.
  • the confidence images may be used adjectively to aid in providing a diagnosis, prognosis, and/or predictive classification of a biological sample.
  • the confidence images may be used to drive areas of interest for micro-dissection of a biological sample. For example, regions of interest identified in the confidence image may be used to identify changes in the gene expression of a biological sample.
  • Micro-dissection may allow for isolated testing and/or micro-detection analysis of regions of a biological sample, for example.
  • micro-detection analysis may be used, for example, to perform gene expression analysis, gene sequencing, Molecular Analysis (e.g., Next Generation Sequencing (NGS)), and targeted therapies for a patient, for example.
  • NGS Next Generation Sequencing
  • the method may include applying a heuristic to an image set 402.
  • a heuristic may include any logical rule that may identify data in a biological sample for micro-dissection.
  • the heuristic may identify areas of the biological sample with low confidence, areas of the biological sample with multiple attributes, areas classified by a certain tissue type, areas classified with a particular class or sub-class of cancer, and clinical data associated with the biological sample.
  • the system may apply one or more heuristics to an image set to identify data in a biological sample that may be of interest for micro-dissection.
  • the method may include receiving a selection of a sample area of the biological sample for removal corresponding to a region of interest 404 and 406.
  • a region of interest may be identified using the results of the heuristic(s) applied to the image sets.
  • the system may receive a selection of the sample area for removal from a medical practitioner.
  • the medical practitioner may use an interface on the system to highlight or otherwise identify a sample area. For example, the medical practitioner may draw a boundary around the sample area to identify the sample area for micro-dissection. In another example, the medical practitioner may highlight an area of the sample for micro-dissection.
  • the system may automatically select the sample area for removal.
  • the system may receive the data from the heuristics and use the data to automatically select sample areas for removal.
  • the system may highlight or otherwise identify the sample area.
  • Figs. 18A-18D illustrate example sample areas in accordance with an aspect of the present invention.
  • the sample areas may be highlighted with a circle, as illustrated in Figs. 18A and 18B.
  • the sample areas may also be highlighted by grids, as illustrated in Figs. 18C and 18D.
  • the sample areas may be selected based upon confidence levels, as illustrated in Fig. 18C.
  • the regions illustrating a low confidence e.g., a plurality of white dots, may be selected as a sample area, as illustrated in Fig. 18C.
  • the method may also include performing a registration between the region of interest and the SHP image 408.
  • a registration between the region of interest and the SHP image may include, for example, associating the spatial location of the region interest with a test identification of the sample area.
  • the system may associate the spatial location from the SHP image with the region of interest, and store the association in a data repository.
  • the system may be able to associate the gene sequencing analysis performed on the region of interest with the spectral data from the SHP image.
  • the system may be able to identify any patterns or changes in patterns in the gene sequencing, based on the analysis performed.
  • the system may be able to track the tests performed on the sections of the sample.
  • Fig. 17 illustrates an example registration between the clinical image and the SHP image, in accordance with an aspect of the present invention.
  • the system may overlay the SHP image with the clinical image to correlate the spectra from the SHP image with the clinical image.
  • the method may include harvesting material from the selected sample area and performing molecular tests 410.
  • the system may use the selected sample area to direct an automated tool to remove the sample area from the biological sample.
  • the automated tool may direct a laser or other milling apparatus to remove the sample area from the biological sample.
  • the system may perform one or more molecular tests on the sample area.
  • the aspects of the present invention may be applied to any particular cell or tissue class, whether cancerous or non-cancerous.
  • the iterative process the most accurate results may be achieved when the first iteration analyzes the original specimen data set for the broadest cell or tissue class and, with each subsequent iteration, analyzes the resulting specimen data subset for a narrower cell or tissue class.
  • the result of any given iteration may be provided or outputted to indicate which portion of the data is associated with a particular condition. For example, if the first iteration is cancer analysis, the method may proceed to a second iteration of the cancerous data, but may also provide or output information regarding the portion of the data that was found to be non-cancerous.
  • Figure 19 shows various features of an example computer system 1900 for use in conjunction with methods in accordance with aspects of invention.
  • the computer system 1900 is used by a requestor/practitioner or other user 1901 or a representative of the requestor/practitioner or other user 1901 via a terminal 1902, such as a personal computer (PC), minicomputer, mainframe computer, microcomputer, telephone device, personal digital assistant (PDA), or other device having a processor and input capability.
  • the server model comprises, for example, a PC, minicomputer, mainframe computer, microcomputer, or other device having a processor and a repository for data or that is capable of accessing a repository of data.
  • the server model 1906 may be associated, for example, with an accessibly repository of disease-based data such as training sets and/or algorithms for use in diagnosis, prognosis and/or predictive analysis.
  • Any of the above-described data may be transmitted between the practitioner and SHP system (or other user), for example, via a network, 1910, such as the Internet, for example, and is transmitted between the analyst 1901 and the server model 1906. Communications are made, for example, via couplings 191 1 , 1913, such as wired, wireless, or fiberoptic links.
  • aspects of the invention may be implemented using hardware, software or a combination thereof and may be implemented in one or more computer systems or other processing systems. In one variation, aspects of the invention are directed toward one or more computer systems capable of carrying out the functionality described herein. An example of such a computer system 2000 is shown in Figure 20.
  • Computer system 2000 includes one or more processors, such as processor 2004.
  • the processor 2004 is connected to a communication infrastructure 2006 (e.g., a communications bus, cross-over bar, or network).
  • a communication infrastructure 2006 e.g., a communications bus, cross-over bar, or network.
  • Computer system 2000 can include a display interface 2002 that forwards graphics, text, and other data from the communication infrastructure
  • Computer system 2000 also includes a main memory 2008, preferably random access memory (RAM), and may also include a secondary memory
  • main memory 2008 preferably random access memory (RAM)
  • RAM random access memory
  • the secondary memory 2010 may include, for example, a hard disk drive 2012 and/or a removable storage drive 2014, representing a floppy disk drive, a magnetic tape drive, an optical disk drive, etc.
  • the removable storage drive 2014 reads from and/or writes to a removable storage unit 2018 in a well-known manner.
  • Removable storage unit 2018 represents a floppy disk, magnetic tape, optical disk, etc., which is read by and written to removable storage drive 2014.
  • the removable storage unit 2018 includes a computer usable storage medium having stored therein computer software and/or data.
  • secondary memory 2010 may include other similar devices for allowing computer programs or other instructions to be loaded into computer system 2000.
  • Such devices may include, for example, a removable storage unit 2022 and an interface 2020. Examples of such may include a program cartridge and cartridge interface (such as that found in video game devices), a removable memory chip (such as an erasable programmable read only memory (EPROM), or programmable read only memory (PROM)) and associated socket, and other removable storage units 2022 and interfaces 2020, which allow software and data to be transferred from the removable storage unit 2022 to computer system 2000.
  • a program cartridge and cartridge interface such as that found in video game devices
  • EPROM erasable programmable read only memory
  • PROM programmable read only memory
  • Computer system 2000 may also include a communications interface 2024.
  • Communications interface 2024 allows software and data to be transferred between computer system 2000 and external devices. Examples of communications interface 2024 may include a modem, a network interface (such as an Ethernet card), a communications port, a Personal Computer Memory Card International Association (PCMCIA) slot and card, etc.
  • Software and data transferred via communications interface 2024 are in the form of signals 2028, which may be electronic, electromagnetic, optical or other signals capable of being received by communications interface 2024. These signals 2028 are provided to communications interface 2024 via a communications path (e.g., channel) 2026.
  • This path 2026 carries signals 2028 and may be implemented using wire or cable, fiber optics, a telephone line, a cellular link, a radio frequency (RF) link and/or other communications channels.
  • RF radio frequency
  • the terms "computer program medium” and “computer usable medium” are used to refer generally to media such as a removable storage drive 2014, a hard disk installed in hard disk drive 2012, and signals 2028. These computer program products provide software to the computer system 2000. Aspects of the invention are directed to such computer program products.
  • Computer programs are stored in main memory 2008 and/or secondary memory 2010. Computer programs may also be received via communications interface 2024. Such computer programs, when executed, enable the computer system 2000 to perform the features in accordance with aspects of the invention, as discussed herein. In particular, the computer programs, when executed, enable the processor 2004 to perform such features. Accordingly, such computer programs represent controllers of the computer system 2000.
  • aspects of the invention are implemented using software
  • the software may be stored in a computer program product and loaded into computer system 2000 using removable storage drive 2014, hard drive 2012, or communications interface 2024.
  • the control logic when executed by the processor 2004, causes the processor 2004 to perform the functions as described herein.
  • aspects of the invention are implemented primarily in hardware using, for example, hardware components, such as application specific integrated circuits (ASICs). Implementation of the hardware state machine so as to perform the functions described herein will be apparent to persons skilled in the relevant art(s).
  • aspects of the invention are implemented using a combination of both hardware and software.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Biomedical Technology (AREA)
  • Surgery (AREA)
  • Veterinary Medicine (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Pathology (AREA)
  • Animal Behavior & Ethology (AREA)
  • Biophysics (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Investigating Or Analysing Biological Materials (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)
  • Image Processing (AREA)

Abstract

La présente invention porte sur des procédés, des dispositifs et des systèmes qui peuvent permettre à un praticien d'obtenir des informations relatives à un échantillon biologique, comprenant des données d'analyse, un diagnostic médical et/ou un pronostic ou une analyse prédictive. Le procédé, les dispositifs et les systèmes peuvent fournir un degré ou un niveau de développement des maladies identifiées. De plus, les procédés, les dispositifs et les systèmes peuvent générer une valeur de confiance pour les classifications prédictives générées, qui peut, par exemple, être générée dans un format pour montrer une telle valeur de confiance ou une autre caractéristique dans une représentation graphique (par exemple, un code couleur). De plus, les procédés, dispositifs et systèmes peuvent aider à l'identification et à la découverte de nouvelles classes et sous-types de tissus.
PCT/US2014/031254 2013-03-19 2014-03-19 Procédé et système d'analyse de spécimens biologiques par imagerie spectrale WO2014153423A2 (fr)

Priority Applications (9)

Application Number Priority Date Filing Date Title
MX2015013415A MX2015013415A (es) 2013-03-19 2014-03-19 Metodo y sistema para analizar especímenes biológicos mediante formación de imágenes espectrales.
KR1020157030157A KR20160012114A (ko) 2013-03-19 2014-03-19 스펙트럼 이미지화에 의해 생물학적 시료를 분석하기 위한 방법 및 시스템
AU2014235921A AU2014235921A1 (en) 2013-03-19 2014-03-19 Method and system for analyzing biological specimens by spectral imaging
BR112015024134A BR112015024134A2 (pt) 2013-03-19 2014-03-19 método para classificar a amostra biológica, sistema para classificar a amostra biológica e produto de programa de computador
JP2016504352A JP2016514869A (ja) 2013-03-19 2014-03-19 スペクトル画像による生物試料の分析方法およびシステム
EP14768134.0A EP2976735A4 (fr) 2013-03-19 2014-03-19 Procédé et système d'analyse de spécimens biologiques par imagerie spectrale
CA2907405A CA2907405A1 (fr) 2013-03-19 2014-03-19 Procede et systeme d'analyse de specimens biologiques par imagerie spectrale
IL241496A IL241496B (en) 2013-03-19 2015-09-10 System and method for analyzing samples using spectral imaging
HK16108997.8A HK1221051A1 (zh) 2013-03-19 2016-07-27 通過光譜成像來分析生物標本的方法和系統

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201361803376P 2013-03-19 2013-03-19
US61/803,376 2013-03-19

Publications (2)

Publication Number Publication Date
WO2014153423A2 true WO2014153423A2 (fr) 2014-09-25
WO2014153423A3 WO2014153423A3 (fr) 2015-11-12

Family

ID=51581798

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2014/031254 WO2014153423A2 (fr) 2013-03-19 2014-03-19 Procédé et système d'analyse de spécimens biologiques par imagerie spectrale

Country Status (10)

Country Link
EP (1) EP2976735A4 (fr)
JP (1) JP2016514869A (fr)
KR (1) KR20160012114A (fr)
AU (1) AU2014235921A1 (fr)
BR (1) BR112015024134A2 (fr)
CA (1) CA2907405A1 (fr)
HK (1) HK1221051A1 (fr)
IL (1) IL241496B (fr)
MX (1) MX2015013415A (fr)
WO (1) WO2014153423A2 (fr)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020232363A1 (fr) * 2019-05-16 2020-11-19 PAIGE.AI, Inc. Systèmes et procédés de traitement d'images visant à classifier les images traitées pour une pathologie numérique
CN112689763A (zh) * 2018-09-20 2021-04-20 美国西门子医学诊断股份有限公司 用于样本分类的假设与验证网络和方法
CN113092403A (zh) * 2021-04-12 2021-07-09 桂林电子科技大学 一种学生校服纺织品材料太赫兹光谱定性检测方法
US11062800B2 (en) 2018-11-09 2021-07-13 Lunit Inc. Method for managing annotation job, apparatus and system supporting the same
CN113257413A (zh) * 2021-06-22 2021-08-13 安翰科技(武汉)股份有限公司 基于深度学习的癌症预后生存预测方法、设备及存储介质

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
AU2018282883B2 (en) * 2017-06-12 2024-03-28 Protein Dynamic Solutions, Inc. Method and system for analysis of crystals and crystallization
EP3844781A4 (fr) * 2018-08-30 2022-05-11 Applied Materials, Inc. Système pour la détection et la classification automatiques de tumeurs
JP7209968B2 (ja) * 2019-10-09 2023-01-23 富士フイルム株式会社 画像処理装置、画像処理装置の作動方法、画像処理装置の作動プログラム
KR102572517B1 (ko) * 2021-08-18 2023-08-31 주식회사 아폴론 인공지능 기반 호산구의 라만 데이터 처리 방법 및 장치

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7136518B2 (en) * 2003-04-18 2006-11-14 Medispectra, Inc. Methods and apparatus for displaying diagnostic data
US7282723B2 (en) * 2002-07-09 2007-10-16 Medispectra, Inc. Methods and apparatus for processing spectral data for use in tissue characterization
JPWO2010123043A1 (ja) * 2009-04-21 2012-10-25 日本電気株式会社 癌の評価方法
CN103096786A (zh) * 2010-05-03 2013-05-08 国际科学技术医疗系统有限责任公司 宫颈瘤变检测和诊断的图像分析
AU2011270731A1 (en) * 2010-06-25 2013-02-07 Cireca Theranostics, Llc Method for analyzing biological specimens by spectral imaging
US9025850B2 (en) * 2010-06-25 2015-05-05 Cireca Theranostics, Llc Method for analyzing biological specimens by spectral imaging
EP2611941A4 (fr) * 2010-08-30 2014-01-22 Myriad Genetics Inc Signatures génétiques pour le diagnostic et le pronostic du cancer
US8711149B2 (en) * 2010-12-08 2014-04-29 Definiens Ag Graphical user interface for interpreting the results of image analysis
WO2013028926A2 (fr) * 2011-08-23 2013-02-28 The Methodist Hospital Research Institute Système d'imagerie par diffusion raman anti-stokes, cohérente, hautement spécifique, basée sur les connaissances et sans marqueur, et procédés associés

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112689763A (zh) * 2018-09-20 2021-04-20 美国西门子医学诊断股份有限公司 用于样本分类的假设与验证网络和方法
US11062800B2 (en) 2018-11-09 2021-07-13 Lunit Inc. Method for managing annotation job, apparatus and system supporting the same
US11335455B2 (en) 2018-11-09 2022-05-17 Lunit Inc. Method for managing annotation job, apparatus and system supporting the same
JP7100336B2 (ja) 2019-05-16 2022-07-13 ペイジ．エーアイ インコーポレイテッド デジタル病理学のために、画像を処理し、処理された画像を分類するためのシステムおよび方法
US11042807B2 (en) 2019-05-16 2021-06-22 PAIGE.AI, Inc. Systems and methods for processing images to classify the processed images for digital pathology
CN114072879A (zh) * 2019-05-16 2022-02-18 佩治人工智能公司 用于处理图像以对经处理的用于数字病理的图像进行分类的系统和方法
US11315029B2 (en) 2019-05-16 2022-04-26 PAIGE.AI, Inc. Systems and methods for processing images to classify the processed images for digital pathology
US10891550B2 (en) 2019-05-16 2021-01-12 PAIGE.AI, Inc. Systems and methods for processing images to classify the processed images for digital pathology
JP2022528014A (ja) * 2019-05-16 2022-06-07 ペイジ．エーアイ インコーポレイテッド デジタル病理学のために、画像を処理し、処理された画像を分類するためのシステムおよび方法
WO2020232363A1 (fr) * 2019-05-16 2020-11-19 PAIGE.AI, Inc. Systèmes et procédés de traitement d'images visant à classifier les images traitées pour une pathologie numérique
US11593684B2 (en) 2019-05-16 2023-02-28 PAIGE.AI, Inc. Systems and methods for processing images to classify the processed images for digital pathology
CN114072879B (zh) * 2019-05-16 2023-09-19 佩治人工智能公司 用于处理图像以对经处理的用于数字病理的图像进行分类的系统和方法
US11893510B2 (en) 2019-05-16 2024-02-06 PAIGE.AI, Inc. Systems and methods for processing images to classify the processed images for digital pathology
CN113092403A (zh) * 2021-04-12 2021-07-09 桂林电子科技大学 一种学生校服纺织品材料太赫兹光谱定性检测方法
CN113092403B (zh) * 2021-04-12 2022-07-01 桂林电子科技大学 一种学生校服纺织品材料太赫兹光谱定性检测方法
CN113257413A (zh) * 2021-06-22 2021-08-13 安翰科技(武汉)股份有限公司 基于深度学习的癌症预后生存预测方法、设备及存储介质

Also Published As

Publication number Publication date
MX2015013415A (es) 2016-08-18
IL241496A0 (en) 2015-11-30
KR20160012114A (ko) 2016-02-02
WO2014153423A3 (fr) 2015-11-12
EP2976735A4 (fr) 2016-12-14
BR112015024134A2 (pt) 2017-07-18
HK1221051A1 (zh) 2017-05-19
EP2976735A2 (fr) 2016-01-27
JP2016514869A (ja) 2016-05-23
AU2014235921A1 (en) 2015-10-15
CA2907405A1 (fr) 2014-09-25
IL241496B (en) 2020-02-27

Similar Documents

Publication Publication Date Title
US9798918B2 (en) Method and system for analyzing biological specimens by spectral imaging
JP6333326B2 (ja) スペクトルイメージングによる生体試料の分析方法
US10043054B2 (en) Methods and systems for classifying biological samples, including optimization of analyses and use of correlation
US20130089248A1 (en) Method and system for analyzing biological specimens by spectral imaging
US9495745B2 (en) Method for analyzing biological specimens by spectral imaging
CA2803933C (fr) Procede pour l'analyse d'echantillons biologiques par imagerie spectrale
EP2976735A2 (fr) Procédé et système d'analyse de spécimens biologiques par imagerie spectrale
EP3729369A2 (fr) Approche de pathologie computationnelle pour une analyse rétrospective d'études d'essais cliniques reposant sur le diagnostic d'accompagnement basés sur des tissus
AU2017204736A1 (en) Method for analyzing biological specimens by spectral imaging
US10460439B1 (en) Methods and systems for identifying cellular subtypes in an image of a biological specimen

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14768134

Country of ref document: EP

Kind code of ref document: A2

ENP Entry into the national phase

Ref document number: 2907405

Country of ref document: CA

WWE Wipo information: entry into national phase

Ref document number: 241496

Country of ref document: IL

ENP Entry into the national phase

Ref document number: 2016504352

Country of ref document: JP

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: MX/A/2015/013415

Country of ref document: MX

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2014768134

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2014235921

Country of ref document: AU

Date of ref document: 20140319

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20157030157

Country of ref document: KR

Kind code of ref document: A

REG Reference to national code

Ref country code: BR

Ref legal event code: B01A

Ref document number: 112015024134

Country of ref document: BR

ENP Entry into the national phase

Ref document number: 112015024134

Country of ref document: BR

Kind code of ref document: A2

Effective date: 20150918