CN112508952B - Pathological section double-objective lens self-adaptive scanning control method and system - Google Patents
Pathological section double-objective lens self-adaptive scanning control method and system Download PDFInfo
- Publication number
- CN112508952B CN112508952B CN202110158033.9A CN202110158033A CN112508952B CN 112508952 B CN112508952 B CN 112508952B CN 202110158033 A CN202110158033 A CN 202110158033A CN 112508952 B CN112508952 B CN 112508952B
- Authority
- CN
- China
- Prior art keywords
- image
- scanning
- slice
- objective lens
- magnification
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 62
- 230000001575 pathological effect Effects 0.000 title claims abstract description 60
- 238000013507 mapping Methods 0.000 claims abstract description 30
- 238000003860 storage Methods 0.000 claims abstract description 29
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000010586 diagram Methods 0.000 claims description 28
- 230000003044 adaptive effect Effects 0.000 claims description 21
- 238000013527 convolutional neural network Methods 0.000 claims description 17
- 238000009826 distribution Methods 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 230000007170 pathology Effects 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 description 22
- 239000013598 vector Substances 0.000 description 10
- 238000003745 diagnosis Methods 0.000 description 6
- 238000003384 imaging method Methods 0.000 description 6
- 238000013135 deep learning Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000010827 pathological analysis Methods 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 101100437734 Nocardia asteroides bla gene Proteins 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
- G01N2021/0106—General arrangement of respective parts
- G01N2021/0112—Apparatus in one mechanical, optical or electronic block
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10141—Special mode during image acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Pathology (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Computing Systems (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Quality & Reliability (AREA)
- Evolutionary Biology (AREA)
- Radiology & Medical Imaging (AREA)
- Medical Informatics (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Multimedia (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
Abstract
The invention discloses a pathological section double-objective lens self-adaptive scanning control method and a pathological section double-objective lens self-adaptive scanning control system, wherein the method comprises the following steps: s01, controlling to scan all areas of a target slice by using a first-magnification objective lens to obtain a slice panoramic image, and storing the obtained slice panoramic image according to a pyramid image storage format; s02, acquiring a slice panoramic image from a multiresolution pyramid image layer for image processing, and identifying and positioning an interested region; and S03, mapping the position coordinates of the region of interest to determine a scanning motion path, and controlling to use a second-magnification objective lens to perform local scanning on the identified region of interest to obtain a corresponding local image, wherein the magnification of the second-magnification objective lens is higher than that of the first-magnification objective lens. The invention can realize the self-adaptive scanning control of the pathological section double-objective lens, and has the advantages of simple realization method, low cost and power consumption, high intelligent degree, and the like, and can give consideration to both the scanning efficiency and the image resolution.
Description
Technical Field
The invention relates to the technical field of digital slice scanning, in particular to a pathological section double-objective lens self-adaptive scanning control method and system.
Background
In the field of pathology, digital slice scan analysis systems are key to converting physical slices in the real world into digital slices.
In order to realize automatic scanning analysis of pathological sections, pathological section scanning analyzers are usually used at present, namely, pathological sections are placed in the scanning analyzers, and the scanning analyzers perform full-area scanning on the sections. However, such pathological section scanning analyzers can only scan the whole area of a section by using an objective lens with a fixed magnification, and thus there is a contradiction between scanning efficiency, scanning magnification and system resources, i.e., it is difficult to consider scanning efficiency, scanning magnification and system resources at the same time. If higher scan efficiency is to be achieved, it is usually difficult to achieve a higher scan magnification, and if a higher scan magnification is required, it is usually impossible to achieve a higher scan efficiency. The digital scanner can achieve higher magnification, but has the problems of high storage space and high cost caused by high-magnification digital imaging.
Not all slice regions of a pathological section are actually of interest to the analyst, who is usually only interested in the areas where pathology-positive regions are present, and thus a sharper image is needed for the areas of interest of the pathology-positive regions. The traditional pathological section scanning analyzer directly performs scanning analysis on the whole area of the whole section, and if an objective lens with lower magnification is directly used, although the whole system resources can be reduced, the obtained whole image of the section is not clear, and the region of interest cannot be clearly analyzed; however, if the objective lens with a higher magnification is directly used, although a clearer image of the whole area can be obtained, the problems of hardware cost and system resource increase are caused.
In order to solve the above problems, practitioners have proposed to provide a plurality of objective lenses in a pathological section scanning analyzer, so that different objective lenses can be switched according to different requirements for scanning. However, in the method, one objective lens is usually selected for scanning in the same scanning process, that is, only images with the same magnification can be obtained in one scanning process, and if a clearer image is required for the same slice, the large-magnification objective lens needs to be manually switched and rescanned again. The method still needs to be switched by manual operation, the scanning process is complex, the scanning efficiency is still low, images with the same magnification ratio can only be obtained in each scanning process, clear images of the required region of interest can be obtained only after continuous switching, the clear images of the required region of interest cannot be rapidly positioned and obtained, and meanwhile, due to the fact that multiple times of scanning are needed, the requirement on system resources of a digital slice scanner is high.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the technical problems in the prior art, the invention provides a pathological section double-objective lens adaptive scanning control method and system which are simple in implementation method, low in cost and power consumption, high in intelligent degree and capable of giving consideration to both scanning efficiency and image resolution.
In order to solve the technical problems, the technical scheme provided by the invention is as follows:
a pathological section double-objective lens adaptive scanning control method comprises the following steps:
step S01 integral scan: controlling to use a first-magnification objective lens to scan all areas of a target slice to obtain a slice panoramic image, and storing the obtained slice panoramic image and the zoomed image of the slice panoramic image according to a pyramid image storage format to form a multi-resolution pyramid image layer;
step S02, region of interest identification: acquiring the slice panoramic image from the multi-resolution pyramid image layer for image processing, and identifying and positioning an interested region;
step S03 local scan: and mapping the position coordinates of the region of interest in the multi-resolution pyramid image layer to actual physical coordinates to be scanned, determining a scanning motion path, controlling to use a second magnification objective lens to perform local scanning on the identified region of interest to obtain a corresponding local image, and storing the local image in a newly constructed image layer in the multi-resolution pyramid image layer according to a coordinate mapping relation to realize sparse storage, wherein the magnification of the second magnification objective lens is higher than that of the first magnification objective lens.
Further, the step of obtaining the slice panorama in step S01 includes:
step S101, preview: controlling the preview slice to obtain the outline of the whole area of the target slice;
step S102, focusing: controlling to focus multiple points on the target slice by using the first magnification objective lens, and fitting to obtain a focusing plane of the whole area of the target slice;
step S103, scanning: scanning the whole area of the target slice according to the fitting result of the focusing plane;
step S104, splicing: and splicing the scanning results obtained in the step S103 to obtain the slice panoramic image.
Further, in step S02, identifying and locating the region of interest by using a deep convolutional neural network, including:
s201, training: training a deep convolutional neural network by using a historical pathology digital panorama database;
s202, predicting: predicting whether each area in the slice panoramic image is a positive area by using a trained deep convolution neural network to obtain the prediction probability distribution of each area;
s203, identifying and positioning: and determining the region of interest according to the prediction probability distribution.
Further, the step S202 includes: acquiring the slice panoramic image stored in the pyramid image storage format in the step S01, wherein the original slice panoramic image is stored in the bottom layer, the original slice panoramic image is stored in each layer except the bottom layer after being zoomed, and the resolution of each layer is gradually reduced from bottom to top; performing foreground segmentation on a top-level picture in the slice panoramic image and then performing binarization to obtain a tissue foreground image; and indexing the image blocks corresponding to the bottom layer image layer in the slice panoramic image according to the pixel coordinates of the tissue foreground image, then sending the image blocks obtained by each index into the trained deep convolutional neural network as input to predict a positive probability value, and finally writing the corresponding positive probability value obtained by predicting the position of each pixel into the corresponding coordinates to form a predicted probability distribution map, namely a thermodynamic diagram.
Further, the step of step S03 includes:
step S301, determining a target area needing to be locally scanned by using a second magnification objective according to the positive probability value in each coordinate range in the region of interest;
step S302, mapping the determined target area into actual physical coordinates to be scanned, and determining a scanning motion path according to the actual physical coordinates obtained by mapping;
and step S303, controlling to use the second magnification objective lens to scan the target area according to the scanning motion path determined in the step S302.
Further, in the step S301, a prediction probability distribution map of the region of interest is obtained, that is, a thermodynamic diagram, and the coordinate ranges in the region of interest are sorted according to the probability according to the thermodynamic diagram; according to a preset positive region probability value threshold valueP thr And a maximum number threshold of positive regionsP num And finally, determining a target area needing to be locally scanned by using the second magnification objective lens.
Further, the step of step S302 includes:
step S321, determining the relative coordinate of the target area relative to the scanning starting point according to the magnification of the second magnification objective;
step S322, mapping the target area into the actual physical coordinate to be scanned according to the relative coordinate;
and step S323, replanning the scanned motion path according to the S-shaped scanning path so as to minimize the distance between the real physical coordinate of the current scanning point and the scanning origin point during each scanning, and determining to obtain the final scanning motion path.
Further, when the multi-resolution pyramid image layer is formed in step S01, specifically, storing an original slice panorama obtained by scanning with the objective lens with the first magnification at the bottom layer, sequentially reducing the original slice panorama according to a specified magnification, and then respectively storing the reduced original slice panorama to each corresponding image layer, wherein the resolution of each image layer is gradually reduced from bottom to top, so as to form an initial multi-resolution pyramid image layer; when the local image is stored in the multiresolution pyramid map layer in step S03, a new bottom layer is constructed in the initial multiresolution pyramid map layer, and the local image is stored in the newly constructed bottom layer according to the coordinate mapping relationship to realize sparse storage, so as to form a final sparse multiresolution pyramid map layer.
Further, the method further comprises the steps of obtaining a selection signal which is input from the outside and used for selecting the region of interest, determining corresponding coordinates under a current display layer in the multi-resolution pyramid image layer according to screen coordinates corresponding to the selection signal, and obtaining and displaying an image according to the determined coordinates.
A pathological section double-objective lens adaptive scanning control system comprises:
the integral scanning module is used for controlling the first-magnification objective lens to scan all areas of a target slice to obtain a slice panoramic image, and storing the obtained slice panoramic image and the zoomed image of the slice panoramic image according to a pyramid image storage format to form a multi-resolution pyramid image layer;
the interested region identification module is used for acquiring the slice panoramic image from the multi-resolution pyramid image layer to perform image processing and identifying and positioning an interested region;
and the local scanning module is used for mapping the position coordinates of the region of interest in the multi-resolution pyramid image layer to actual physical coordinates to be scanned, determining a scanning motion path, controlling a second magnification objective lens to locally scan the identified region of interest to obtain a corresponding local image, and storing the local image in a newly constructed image layer in the multi-resolution pyramid image layer according to a coordinate mapping relation to realize sparse storage, wherein the magnification of the second magnification objective lens is higher than that of the first magnification objective lens.
A pathological section double-objective lens adaptive scanning control system comprises a processor and a memory, wherein the memory is used for storing a computer program, the processor is used for executing the computer program, and the processor is used for executing the computer program so as to execute the method.
Compared with the prior art, the invention has the advantages that:
1. the invention simulates the actual analysis process of pathological sections, rapidly scans all the areas of the sections by using the low-magnification objective lens to obtain a section panoramic image, then performs image processing on the section panoramic image to identify and locate the region of interest, and then performs higher-magnification scanning on the identified and located region of interest by using the higher-magnification objective lens, so that the rapid scanning of the sections can be ensured, the requirements on system resources can be reduced, and simultaneously higher image resolution of the region of interest can be obtained, namely, the requirements on scanning efficiency, image resolution and system resources can be simultaneously considered, the whole scanning process can be free from manual participation, and the intelligent scanning analysis of the pathological sections can be realized.
2. On the basis of using the low-magnification objective lens to carry out integral fast scanning on the slice, the interested pathological area is automatically identified and positioned by combining the image processing and the deep learning method, so that the automatic switching of the high-magnification objective lens and the low-magnification objective lens can be realized in a matching way, the imaging magnification is adaptively adjusted, the observation and diagnosis process of the pathological slice is simulated, the double-objective lens imaging is automatically switched in the whole process, the requirement on system resources can be reduced under the condition of keeping fast scanning, the diagnosis requirement on the high-resolution pathological area is considered, and the pathological diagnosis process can be efficiently assisted and completed.
3. The invention further constructs a sparse multi-resolution pyramid image layer format to support an automatic and external input mode, and stores the high-magnification clear image acquired correspondingly in a sparse mode into the bottommost sparse image layer in the multi-resolution pyramid format, thereby being convenient for storing the local high-resolution image, being easy to display the local high-resolution image by combining a software front-end display interface, and simultaneously reducing the transmission of unnecessary image data, thereby efficiently completing the process from sparse acquisition and sparse storage to final sparse display of the local high-magnification clear image.
4. The invention further combines the high-magnification objective lens and the S-shaped scanning method to realize the quick secondary scanning of the interested region, the distance between the real physical coordinate of the current scanning point and the scanning origin is minimum during each scanning, the high image resolution of the interested region can be ensured, the scanning efficiency can be improved as much as possible, and the scanning efficiency and the comprehensive performance of the image resolution can be further improved.
Drawings
Fig. 1 is a schematic flow chart of an implementation of the pathological section double-objective adaptive scanning control method according to the embodiment.
Fig. 2 is a schematic diagram of the embodiment for implementing the dual-objective adaptive scanning of the pathological section.
Fig. 3 is a schematic diagram of a scanning path for scanning a slice in the present embodiment.
Fig. 4 is a schematic diagram illustrating the principle of storing the scanning result of the low-magnification objective lens in a pyramid format in the present embodiment.
Fig. 5 is a schematic diagram of a flow of manufacturing a data set in the present embodiment.
Fig. 6 is a flowchart illustrating the generation of the thermodynamic diagram in the present embodiment.
Fig. 7 is a schematic diagram illustrating the principle of storing the scanning result of the high-magnification objective lens in a pyramid format in the present embodiment.
Detailed Description
The invention is further described below with reference to the drawings and specific preferred embodiments of the description, without thereby limiting the scope of protection of the invention.
As shown in fig. 1 and 2, the method for controlling the pathological section double-objective adaptive scanning according to the embodiment includes the following steps:
step S01 integral scan: controlling to use a first magnification objective lens to scan all areas of a target slice to obtain a slice panoramic image, and storing the obtained slice panoramic image and the zoomed slice panoramic image according to a pyramid image storage format to form a multi-resolution pyramid image layer;
step S02, region of interest identification: acquiring a slice panoramic image from a multiresolution pyramid image layer for image processing, and identifying and positioning an interested region;
step S03 local scan: and mapping the position coordinates of the region of interest in the multi-resolution pyramid image layer to actual physical coordinates to be scanned, determining a scanning motion path, controlling to use a second magnification objective lens to perform local scanning on the identified region of interest to obtain a corresponding local image, and storing the local image in a newly constructed image layer in the multi-resolution pyramid image layer according to a coordinate mapping relation to realize sparse storage, wherein the magnification of the second magnification objective lens is higher than that of the first magnification objective lens.
In the actual analysis process of pathological sections, an analyst usually uses the low power objective lens to perform rough condition observation, and then uses the high power objective lens to perform careful observation on the region of interest. This embodiment simulates pathological section's actual analysis process, carry out the fast scan to sliced whole region through using low magnification objective earlier, obtain the section panorama, then carry out image processing to the section panorama, the region of interest is fixed out in the discernment, later use higher magnification objective to fix out the region of interest to the discernment again and carry out the scan of higher magnification, can realize objective self-adaptation regulation control, make both can guarantee sliced fast scan, can reduce the requirement to system resources again, can also compromise simultaneously and acquire the higher image resolution ratio of region of interest, can compromise simultaneously scanning efficiency, the requirement of image resolution ratio and system resources, and whole scanning process can need not artifical the participation, can realize pathological section's intelligent scanning analysis.
In the embodiment, the low-magnification objective lens is firstly adopted to rapidly scan the whole area of the whole slice, and the low-magnification objective lens is adopted in the process, so that the requirements on the precision and the like of the focusing module are not high, and the hardware cost can be effectively reduced.
In this embodiment, the step of obtaining the slice panorama in step S01 includes:
step S101, preview: controlling the preview slice to obtain the outline of the whole area of the target slice;
step S102, focusing: controlling to focus multiple points on the target slice by using a first-magnification objective lens, and fitting to obtain a focusing plane of the whole area of the target slice;
step S103, scanning: scanning the whole area of the target slice according to the fitting result of the focusing plane;
step S104, splicing: and splicing the scanning results obtained in the step S103 to obtain a slice panoramic image.
The above step S101 may be specifically implemented by using a preview camera to obtain the outline of the whole region when previewing the slice. The fast scan path can be determined by previewing the slice, and specifically, the fast scan path can be determined by the following method: avoid scanning from the top left origin of the slice, while from the starting point (x) with the foreground of the tissue image, based on the contour of the whole area0, y0) Scanning is started and the width W and height H of the scan are confirmed.
In step S102, after the first magnification objective lens is used to perform fast focusing on multiple points on the wafer, a polynomial fitting manner is used to fit and obtain a focusing plane of the entire region. The low-power objective lens is adopted in the focusing process, so that the requirements on the precision and the like of the focusing module are not high, and the hardware cost and the requirements on system resources can be reduced.
In the step S103, the whole area to be scanned is divided into W × H areas, and the motor moves according to the S-type scanning method, so as to obtain the scanning path as shown in fig. 3; and finishing the scanning process of the whole area of the slice according to the fitting result of the focusing plane and the preset focus value of each scanning area. The combination of the low-magnification objective lens and the S-shaped scanning method can realize the rapid scanning of the whole area of the slice.
In a specific application embodiment, the detailed steps of the overall scanning performed in step S01 are as follows: presetting relevant initial parameters of first-magnification objective scanning, and controlling a motor to start scanning by a system; previewing the slice with a preview camera, obtaining the outline of the whole area, and analyzing the image to obtain a fast scan path to confirm the starting point (x) from the foreground of the tissue image0, y0) Starting scanning, and confirming the width W and the height H of the scanning;fast focusing multiple points on the wafer by using a first-magnification objective lens, and obtaining a focusing plane of the whole area through polynomial fitting; dividing the whole area to be scanned into W multiplied by H areas, and moving a motor according to an S-shaped scanning method to obtain a scanning path; according to the result of the focusing plane fitting and the preset focus value of each scanning area, completing the scanning process of the whole area of the slice; and obtaining a digital pathological section panoramic image under the low-magnification objective lens after splicing.
In this embodiment, after the scanning in step S01, the sliced panorama is stored in a multi-resolution pyramid image storage format, where the multi-resolution pyramid image format is as shown in fig. 4, the original sliced panorama is stored in the bottommost layer, the original sliced panorama is sequentially reduced according to a specified magnification and then is stored in each layer except the bottommost layer, and the resolution of each layer is gradually reduced from bottom to top. By the multi-resolution pyramid image storage format, multi-resolution image data can be obtained, image display under different requirements can be conveniently met, and transmission of unnecessary image data can be reduced.
It is understood that, in step S01, the low-power objective lens may be controlled to adopt other scanning manners according to actual requirements, so as to perform overall fast scanning on the slice.
In step S02, the region of interest is specifically identified and located by using a deep convolutional neural network, so as to realize intelligent and accurate location of the pathological region of interest by using a deep learning method.
The specific steps of identifying and positioning the region of interest by adopting the deep convolutional neural network in the embodiment comprise:
s201, training: training a deep convolutional neural network by using a historical pathology digital panorama database;
s202, predicting: predicting whether each area in the slice panoramic image is a positive area by using the trained deep convolution neural network to obtain the prediction probability distribution of each area;
s203, identifying and positioning: and determining the region of interest according to the predicted probability distribution.
In this embodiment, a large number of historical pathological digital panoramas are used for data set production, as shown in fig. 5, a large number of historical pathological digital panoramas are collected, each pathological digital panoramas are stored according to a multi-resolution pyramid format, a top layer picture with low resolution is taken out for image processing, a tissue foreground picture is obtained and binarized to discard blank picture blocks, corresponding picture blocks located in bottom layer layers of the pyramid are indexed according to pixel coordinates of the binarized foreground picture, and then each picture block is cut into a corresponding data set. And training the deep convolutional neural network by using the manufactured data set, and testing the trained deep convolutional neural network to finally obtain the optimized deep convolutional neural network which accords with the required identification and positioning precision. The deep convolutional neural network has the function of predicting the probability of positive areas in the slice.
In the embodiment, a trained and optimized deep convolutional neural network is adopted to carry out pathological prediction on the whole region of the digital slice panoramic image acquired in the step S01 to obtain a positive region prediction probability distribution map, namely a thermodynamic map, which represents the probability of whether each image block is a positive region, wherein the deeper the color is, the higher the probability of the corresponding positive region is; as shown in fig. 6, a region of interest can be identified and located by using a thermodynamic diagram, a top-layer picture of the slice panorama is taken out for foreground segmentation, binarization is performed, a binarized tissue foreground map subjected to foreground segmentation is obtained, a blank background is discarded, unnecessary processing data amount is reduced, a corresponding image block of a bottom-layer picture layer of the slice panorama is indexed according to pixel coordinates of the foreground map, then the image block obtained by each index is input into a trained and optimized network for positive probability value prediction, and finally the corresponding positive probability value obtained by each pixel position prediction is written into corresponding coordinates to obtain the thermodynamic diagram.
On the basis of using the low-magnification objective lens to carry out overall fast scanning on the slices, the method is combined with image processing and deep learning methods, the interested pathological area is automatically identified and positioned, further local scanning can be realized by the identified and positioned interested area, automatic switching of the high-magnification objective lens and the low-magnification objective lens can be realized in a matched mode, the diagnosis process from overall pathological conditions to local high-definition pathological analysis is seen by simulation analysts from the low-magnification objective lens, full-flow automatic switching double-objective imaging is realized, under the condition of keeping fast scanning, the requirements on system resources are reduced, and meanwhile, the diagnosis requirements on the high-resolution pathological area are considered.
It is understood that other methods can be adopted to realize the identification and positioning of the region of interest according to actual requirements.
In this embodiment, after the region of interest is located, the high-magnification objective lens is used to perform local scanning on the pathological region of interest, so as to obtain a high-magnification clear pathological image. The specific steps of step S03 in this embodiment include:
step S301, determining a target area needing to be locally scanned by using a second magnification objective according to the positive probability value in each coordinate range in the region of interest;
step S302, mapping the determined target area into actual physical coordinates to be scanned, and determining a scanning motion path according to the actual physical coordinates obtained by mapping;
and step S303, controlling to use the second magnification objective lens to scan the target area according to the scanning motion path determined in the step S302.
In step S301, obtaining a prediction probability distribution map of the region of interest, that is, a thermodynamic diagram, and sorting coordinate ranges in the region of interest according to probability according to the thermodynamic diagram; according to a preset positive region probability value threshold valueP thr And a maximum number threshold of positive regionsP num And finally, determining a target area needing to be locally scanned by using the second magnification objective lens, so that local high-magnification scanning can be ensured only for the interested positive area.
In a specific application embodiment, the detailed step of step S301 includes:
when locating a pathological region of interest according to thermodynamic diagrams, the size of the thermodynamic diagrams corresponds to the size of the top-level picture in the slice panorama, and the thermodynamic diagrams are converted into one-dimensional positive region probability vectors containing coordinate information by pixel coordinates, which can be expressed as:
wherein the content of the first and second substances,
representing pixels in a thermodynamic diagramxCoordinates andya positive probability value vector in the coordinates of,xis shown in the range ofx 1 ~x M ，MIs the number of pixel columns in the thermodynamic diagram,yis shown in the range ofy 1 ~ y N，NFor the number of pixel rows in the thermodynamic diagram,
representing a vector transpose;
sequencing the interested regions according to the positive probability obtained by estimation from big to small, and obtaining a one-dimensional positive region probability vector after sequencing, namely:
wherein the content of the first and second substances,
representing pixel coordinates after sorting from large to small according to positive region probability values
And
the following positive probability value vector is used,
indicated range the same asxThe coordinates of the position of the object to be imaged,
indicated range the same asyCoordinates;
according to the threshold value of the probability value of the preset positive areaP thr And a maximum number threshold of positive regionsP num And determining a positive area needing to be scanned by the objective lens with the second magnification, wherein the positive area can be specifically configured as follows: the probability value of the determined positive area is larger thanP thr And the maximum number is not more thanP num (ii) a The final area coordinate vector needing high-magnification objective lens scanning is obtained as follows:
whereink 1Andk 2is shown asKUnder one coordinate
And
the index is set under the coordinate, and the index,
。
through the steps, the positive regions needing important attention can be automatically screened out according to the probability distribution of the positive regions, and the condition that unimportant regions are scanned with high magnification is avoided.
In this embodiment, the step S302 includes:
step S321, determining the relative coordinate of the target area relative to the scanning starting point according to the multiplying power of the second multiplying power objective;
step S322, mapping the target area into the actual physical coordinate to be scanned according to the relative coordinate;
and step S323, replanning the scanned motion path according to the S-shaped scanning path so as to minimize the distance between the real physical coordinate of the current scanning point and the scanning origin point during each scanning, and determining to obtain the final scanning motion path.
In the above steps of this embodiment, the second magnification objective lens and the S-type scanning method may be combined to implement fast secondary high-magnification scanning on the region of interest, and the distance between the real physical coordinate of the current scanning point and the scanning origin is the smallest during each scanning, so that not only the high image resolution of the region of interest can be ensured, but also the scanning efficiency can be improved as much as possible, and the scanning efficiency and the comprehensive performance of the image resolution can be further improved.
In a specific application embodiment, the detailed step of step S302 is:
firstly, according to the determined coordinate vector of the scanning area of the high-magnification objective lensXY K Mapping to real physical coordinates to be scanned actually, wherein the widths and heights of real scanning areas to be scanned obtained by shooting with the preview camera are W and H as described in step S01, and correspond to the column number and row number of preview images obtained by shooting with the preview camera, such as N and M in step S01; the width and height of the pixel size (i.e. the width and height of each pixel) of the camera are n and m respectively, and the magnification of the high-power objective lens is B, so that the camera corresponds to the area coordinate in the thermodynamic diagram
The relative abscissa and ordinate corresponding to the scanning start point are obtained as:
then, a slice origin corresponding to the motor motion is obtained, and the real motor motion coordinate is:
a series of real motion coordinates can be obtained in the above manner as follows:
further replanning the motor motion path according to the scanning method of the motor S, taking the chessboard distance D from the coordinates to the origin as an index, wherein the chessboard distance D represents two coordinates
And
the sum of absolute values of the coordinate value differences of (a) is:
when the motor motion path is re-planned, the motor motion path is firstly planned
Minimum coordinates of D value from origin
Put it in the first place, then find the remaining coordinates and
and then the coordinate with the smallest D value
The motion coordinate vector is finally planned according to the motor motion path by analogy, and the distance between the real physical coordinate of the current scanning point and the scanning origin can be ensured to be minimum during each scanning according to the planned motion path, so that the scanning is improved as much as possibleEfficiency. The motion coordinate vector obtained by final planning is specifically as follows:
then according to the planned motion coordinate vector
And the system controls the motor to rapidly move to each positioned interested area, and uses the second high-magnification objective lens to scan the local high-magnification objective lens.
In this embodiment, after the step S03, the local high-magnification clear image obtained by scanning is stored in the multi-resolution pyramid image layer again, and during storage, the local image is stored in the newly constructed bottommost layer of the multi-resolution pyramid image layer according to the coordinate mapping relationship, so as to implement sparse storage, thereby forming a sparse multi-resolution pyramid image layer. Firstly, a bottommost layer is constructed in the original multi-resolution pyramid image layers, and a position obtained based on analysis and positioning is constructed
Corresponding relation exists with original thermodynamic diagram coordinate positionXY K According to the corresponding motion coordinates
Find the corresponding thermodynamic diagramiA coordinate
And then mapping the pixel coordinates of the thermal image to the bottommost layer to obtain the position of the local high-magnification clear image at the bottommost layer, correspondingly cutting the local high-magnification clear image into tile image blocks, and storing the tile image blocks into the position corresponding to the bottommost layer to obtain the sparse multi-resolution pyramid image storage format. As shown in fig. 7, a slice panorama originally obtained by scanning with a first-magnification objective lens is stored according to a multi-resolution pyramid format to form an initial multi-resolution pyramid image layer; to is coming toAnd storing a local image obtained by scanning the second magnification objective lens, constructing a new bottommost layer on the bottommost layer of the initial multi-resolution pyramid image layer, and storing the local image obtained by scanning and acquiring the second magnification objective lens into the newly-built bottommost layer image layer according to the mapping relation of thermodynamic diagram coordinates to form a sparse multi-resolution pyramid image storage format. By adopting the storage format, the storage of the local high-resolution image can be facilitated, the local high-resolution image is easy to display by combining a software front-end display interface, and meanwhile, the transmission of unnecessary image data can be reduced.
In a specific application embodiment, software operation corresponding to the storage format of the sparse multi-resolution pyramid image capable of being displayed can be correspondingly realized at the display front end in a matching manner according to the self-defined storage format of the sparse multi-resolution pyramid image, so that the final double-objective lens adaptive magnification switching scanning analysis process is completed.
According to the invention, through intelligently positioning a pathological region of interest, adaptively adjusting imaging magnification, simulating an observation and diagnosis process of pathological sections, firstly scanning all regions of the whole pathological section by using a low-magnification objective lens, then quickly positioning an interested pathological region by using a deep learning method, and finally locally scanning the interested pathological region obtained by quick positioning by using a high-magnification objective lens, the requirements on system resources can be reduced under the condition of keeping quick scanning, meanwhile, the diagnosis requirements on a high-resolution pathological region are considered, and the pathological diagnosis process can be efficiently completed in an auxiliary manner.
In this embodiment, the method further includes obtaining a selection signal for selecting the region of interest, which is input from the outside, determining a corresponding coordinate of a current display layer in the multi-resolution pyramid image layer according to a screen coordinate corresponding to the selection signal, and obtaining and displaying an image according to the determined coordinate, so that a required image can be flexibly displayed according to the selection signal input from the outside, thereby implementing a manual operation mode.
Specifically, the high-magnification image of other areas needing to be checked can be configured, the mode is switched to a manual mode, and a mouse is used for clicking the imaging area of the low-magnification objective lens to be checked at the front end of the display interface to obtainTaking corresponding screen coordinates
Obtaining the corresponding coordinates of the display layer positioned under the pyramid by utilizing the relative transformation relation between the screen coordinates and the interface coordinates
The number of layers different from the top layer is set ast1, the number of layers differing from the bottom sparse layer ist2, obtaining the coordinates corresponding to the top layer
Coordinates corresponding to the bottom sparse layer
Respectively, as follows:
wherein the content of the first and second substances,
representing a power of 2 k, with scaling between the multi-resolution pyramid layers by 2.
Further, top level coordinates
Correspondingly obtaining the motion coordinate of the motor according to the transformation relation in the step S03
Correspondingly, the clear image acquired by high magnification is stored in the corresponding coordinate of the bottommost sparse layer
In the middle, the storage and updating of the sparse multiresolution pyramid images are completed.
According to the method, the automatic and external input mode can be supported by constructing the sparse multi-resolution pyramid image layer format, the high-magnification clear images acquired correspondingly are stored in the bottommost sparse image layer in the multi-resolution pyramid format in a sparse mode, and corresponding format analysis is carried out at the front end of interface display according to the corresponding self-defined sparse multi-resolution pyramid image layer format, so that the process from sparse acquisition and sparse storage to final sparse display of local high-magnification clear images can be efficiently completed.
This embodiment pathological section double objective self-adaptation scanning control system includes:
the integral scanning module is used for controlling the first-magnification objective lens to scan all areas of the target slice to obtain a slice panoramic image;
the interest area identification module is used for identifying and positioning the interest area in the panoramic picture of the slice;
and the local scanning module is used for controlling the second magnification objective lens to locally scan the identified interested region to obtain a corresponding local image, and the magnification of the second magnification objective lens is higher than that of the first magnification objective lens.
In this embodiment, the integral scanning module includes:
the preview unit is used for controlling the preview slice and acquiring the outline of the whole area of the target slice;
the focusing unit is used for controlling the first-magnification objective lens to focus multiple points on the target slice and fitting to obtain a focusing plane of the whole area of the target slice;
the scanning unit is used for scanning the whole area of the target slice according to the fitting result of the focusing plane;
and the splicing unit is used for splicing the scanning results obtained in the step S103 to obtain a slice panoramic image.
In this embodiment, the region of interest identification module identifies and locates the region of interest by using a deep convolutional neural network, and the region of interest identification module specifically includes:
the training unit is used for training the deep convolutional neural network by using the historical pathology digital panoramic map database;
the prediction unit is used for predicting whether each area in the slice panoramic image is a positive area by using the trained deep convolutional neural network to obtain the prediction probability distribution of each area;
and the identification positioning unit is used for determining the region of interest according to the prediction probability distribution.
In this embodiment, the local scanning module includes:
the first unit is used for determining a target area needing to be locally scanned by using a second magnification objective according to the positive probability value in each coordinate range in the region of interest;
the second unit is used for mapping the determined target area into actual physical coordinates to be scanned and determining a scanning motion path according to the actual physical coordinates obtained by mapping;
and a third unit for controlling the scanning of the target area using the second magnification objective lens according to the scanning motion path determined by the second unit.
In the embodiment, the prediction probability distribution map of the region of interest is obtained in the first unit, namely the thermodynamic diagram, and the coordinate ranges in the region of interest are sorted according to the probability according to the thermodynamic diagram; according to a preset positive region probability value threshold valueP thrAnd a maximum number threshold of positive regionsP num And finally, determining a target area needing to be locally scanned by using the second magnification objective lens.
The pathological section double-objective lens adaptive scanning control system and the pathological section double-objective lens adaptive scanning control method in the embodiment are in one-to-one correspondence, and have the same implementation principle and technical effect, and are not described in detail herein.
In other application embodiments, the pathological section double-objective adaptive scanning control system of the present invention may further include: the system comprises a processor and a memory, wherein the memory is used for storing a computer program, the processor is used for executing the computer program, and the processor is used for executing the computer program so as to execute the pathological section double-objective adaptive scanning control method.
The foregoing is considered as illustrative of the preferred embodiments of the invention and is not to be construed as limiting the invention in any way. Although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical spirit of the present invention should fall within the protection scope of the technical scheme of the present invention, unless the technical spirit of the present invention departs from the content of the technical scheme of the present invention.
Claims (9)
1. A pathological section double-objective lens adaptive scanning control method is characterized by comprising the following steps:
step S01 integral scan: controlling to use a first-magnification objective lens to scan all areas of a target slice to obtain a slice panoramic image, and storing the obtained slice panoramic image and the zoomed image of the slice panoramic image according to a pyramid image storage format to form a multi-resolution pyramid image layer;
step S02, region of interest identification: acquiring the slice panoramic image from the multi-resolution pyramid image layer for image processing, and identifying and positioning an interested region;
step S03 local scan: mapping the position coordinates of the region of interest in the multi-resolution pyramid image layer to actual physical coordinates to be scanned, determining a scanning motion path, controlling to use a second magnification objective lens to perform local scanning on the identified region of interest to obtain a corresponding local image, and storing the local image in a newly constructed image layer in the multi-resolution pyramid image layer according to a coordinate mapping relation to realize sparse storage, wherein the magnification of the second magnification objective lens is higher than that of the first magnification objective lens;
the step of step S03 includes:
step S301, determining a target area needing to be locally scanned by using a second magnification objective according to the positive probability value in each coordinate range in the region of interest;
step S302, mapping the determined target area into actual physical coordinates to be scanned, and determining a scanning motion path according to the actual physical coordinates obtained by mapping;
and step S303, controlling to use the second magnification objective lens to scan the target area according to the scanning motion path determined in the step S302.
2. The pathological section double-objective adaptive scanning control method according to claim 1, wherein the step of obtaining the slice panorama in step S01 includes:
step S101, preview: controlling the preview slice to obtain the outline of the whole area of the target slice;
step S102, focusing: controlling to focus multiple points on the target slice by using the first magnification objective lens, and fitting to obtain a focusing plane of the whole area of the target slice;
step S103, scanning: scanning the whole area of the target slice according to the fitting result of the focusing plane;
step S104, splicing: and splicing the scanning results obtained in the step S103 to obtain the slice panoramic image.
3. The pathological section double-objective adaptive scanning control method according to claim 1, wherein the identifying and positioning of the region of interest in step S02 using a deep convolutional neural network comprises:
s201, training: training a deep convolutional neural network by using a historical pathology digital panorama database;
s202, predicting: predicting whether each area in the slice panoramic image is a positive area by using a trained deep convolution neural network to obtain the prediction probability distribution of each area;
s203, identifying and positioning: and determining the region of interest according to the prediction probability distribution.
4. The pathological section double-objective adaptive scanning control method according to claim 3, wherein the step S202 comprises: acquiring the slice panoramic image stored in the pyramid image storage format in the step S01, wherein the original slice panoramic image is stored in the bottom layer, the original slice panoramic image is stored in each layer except the bottom layer after being zoomed, and the resolution of each layer is gradually reduced from bottom to top; performing foreground segmentation on a top-level picture in the slice panoramic image and then performing binarization to obtain a tissue foreground image; and indexing the image blocks corresponding to the bottom layer image layer in the slice panoramic image according to the pixel coordinates of the tissue foreground image, then sending the image blocks obtained by each index into the trained deep convolutional neural network as input to predict a positive probability value, and finally writing the corresponding positive probability value obtained by predicting the position of each pixel into the corresponding coordinates to form a predicted probability distribution map, namely a thermodynamic diagram.
5. The pathological section double-objective adaptive scanning control method according to claim 1, wherein in step S301, a predicted probability distribution map of the region of interest, that is, a thermodynamic diagram, is obtained, and the coordinate ranges in the region of interest are sorted according to probability according to the thermodynamic diagram; according to a preset positive region probability value threshold valueP thr And a maximum number threshold of positive regionsP num And finally, determining a target area needing to be locally scanned by using the second magnification objective lens.
6. The pathological section double-objective adaptive scanning control method according to claim 1, wherein: the step of step S302 includes:
step S321, determining the relative coordinate of the target area relative to the scanning starting point according to the magnification of the second magnification objective;
step S322, mapping the target area into the actual physical coordinate to be scanned according to the relative coordinate;
and step S323, replanning the scanned motion path according to the S-shaped scanning path so as to minimize the distance between the real physical coordinate of the current scanning point and the scanning origin point during each scanning, and determining to obtain the final scanning motion path.
7. The pathological section double-objective adaptive scanning control method according to any one of claims 1 to 6, characterized in that: when the multiresolution pyramid image layer is formed in step S01, specifically, storing an original slice panorama obtained by scanning with the first-magnification objective lens at the bottommost layer, sequentially reducing the original slice panorama according to a specified magnification ratio, and then respectively storing the reduced original slice panorama to each corresponding image layer, wherein the resolution ratio of each image layer is gradually reduced from bottom to top, so as to form an initial multiresolution pyramid image layer; when the local image is stored in the multiresolution pyramid map layer in step S03, a new bottom layer is constructed in the initial multiresolution pyramid map layer, and the local image is stored in the newly constructed bottom layer according to the coordinate mapping relationship to realize sparse storage, so as to form a final sparse multiresolution pyramid map layer.
8. A pathological section double-objective lens adaptive scanning control system is characterized by comprising:
the integral scanning module is used for controlling the first-magnification objective lens to scan all areas of a target slice to obtain a slice panoramic image, and storing the obtained slice panoramic image and the zoomed image of the slice panoramic image according to a pyramid image storage format to form a multi-resolution pyramid image layer;
the interested region identification module is used for acquiring the slice panoramic image from the multi-resolution pyramid image layer to perform image processing and identifying and positioning an interested region;
the local scanning module is used for mapping the position coordinates of the region of interest in the multi-resolution pyramid image layer to actual physical coordinates to be scanned and then determining a scanning motion path, controlling a second magnification objective lens to locally scan the identified region of interest to obtain a corresponding local image, and storing the local image in a newly constructed image layer in the multi-resolution pyramid image layer according to a coordinate mapping relation to realize sparse storage, wherein the magnification of the second magnification objective lens is higher than that of the first magnification objective lens;
the local scanning module includes:
the first unit is used for determining a target area needing to be locally scanned by using a second magnification objective according to the positive probability value in each coordinate range in the region of interest;
the second unit is used for mapping the determined target area into actual physical coordinates to be scanned and determining a scanning motion path according to the actual physical coordinates obtained by mapping;
a third unit, configured to control to scan the target area according to the scanning motion path determined in step S302 by using the second magnification objective lens.
9. A pathological section double-objective adaptive scanning control system, comprising a processor and a memory, wherein the memory is used for storing a computer program, and the processor is used for executing the computer program to execute the method according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110158033.9A CN112508952B (en) | 2021-02-05 | 2021-02-05 | Pathological section double-objective lens self-adaptive scanning control method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110158033.9A CN112508952B (en) | 2021-02-05 | 2021-02-05 | Pathological section double-objective lens self-adaptive scanning control method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112508952A CN112508952A (en) | 2021-03-16 |
| CN112508952B true CN112508952B (en) | 2021-05-04 |
Family
ID=74952665
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110158033.9A Active CN112508952B (en) | 2021-02-05 | 2021-02-05 | Pathological section double-objective lens self-adaptive scanning control method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112508952B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113237881B (en) * | 2021-05-10 | 2023-09-12 | 博奥生物集团有限公司 | Detection method and device for specific cells and pathological section detection system |
| CN113766131B (en) * | 2021-09-15 | 2022-05-13 | 广州市明美光电技术有限公司 | Multi-target point focusing method and digital slice scanner applying same |
| TWI826924B (en) * | 2022-02-11 | 2023-12-21 | 友達光電股份有限公司 | Dual objective lenses microscope system |
| CN117649381A (en) * | 2023-11-27 | 2024-03-05 | 北京透彻未来科技有限公司 | Device for automatically calculating zoom ratio based on digital pathological section |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101339129A (en) * | 2007-09-03 | 2009-01-07 | 深圳大学 | Variable visual field scanning microscope and the method based on fixed light path system |
| CN101395465A (en) * | 2006-03-03 | 2009-03-25 | 3D希斯特赫有限公司 | Automated digital image recording system for and method of digitizing slides |
| CN102368283A (en) * | 2011-02-21 | 2012-03-07 | 麦克奥迪实业集团有限公司 | Digital slice-based digital remote pathological diagnosis system and method |
| CN102542154A (en) * | 2011-12-02 | 2012-07-04 | 麦克奥迪实业集团有限公司 | Method for carrying out quality control on pathological digital sections on basis of cloud storage |
| CN210803857U (en) * | 2019-09-11 | 2020-06-19 | 宁波江丰生物信息技术有限公司 | Double-objective switching mechanism |
| CN112150446A (en) * | 2020-09-27 | 2020-12-29 | 郑州中普医疗器械有限公司 | Microscope examination and film reading scanning method, computer equipment and system |
| CN112270684A (en) * | 2020-12-25 | 2021-01-26 | 宁波兰茜生物科技有限公司 | Microscopic image immunohistochemical virtual multiple labeling and analyzing method and system |
-
2021
- 2021-02-05 CN CN202110158033.9A patent/CN112508952B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101395465A (en) * | 2006-03-03 | 2009-03-25 | 3D希斯特赫有限公司 | Automated digital image recording system for and method of digitizing slides |
| CN101339129A (en) * | 2007-09-03 | 2009-01-07 | 深圳大学 | Variable visual field scanning microscope and the method based on fixed light path system |
| CN102368283A (en) * | 2011-02-21 | 2012-03-07 | 麦克奥迪实业集团有限公司 | Digital slice-based digital remote pathological diagnosis system and method |
| CN102542154A (en) * | 2011-12-02 | 2012-07-04 | 麦克奥迪实业集团有限公司 | Method for carrying out quality control on pathological digital sections on basis of cloud storage |
| CN210803857U (en) * | 2019-09-11 | 2020-06-19 | 宁波江丰生物信息技术有限公司 | Double-objective switching mechanism |
| CN112150446A (en) * | 2020-09-27 | 2020-12-29 | 郑州中普医疗器械有限公司 | Microscope examination and film reading scanning method, computer equipment and system |
| CN112270684A (en) * | 2020-12-25 | 2021-01-26 | 宁波兰茜生物科技有限公司 | Microscopic image immunohistochemical virtual multiple labeling and analyzing method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112508952A (en) | 2021-03-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112508952B (en) | Pathological section double-objective lens self-adaptive scanning control method and system | |
| CN113537106A (en) | Fish feeding behavior identification method based on YOLOv5 | |
| JP4806230B2 (en) | Deterioration dictionary generation program, method and apparatus | |
| CN112183788A (en) | Domain adaptive equipment operation detection system and method | |
| CN108961180B (en) | Infrared image enhancement method and system | |
| JP2013117848A (en) | Image processing apparatus and image processing method | |
| CN110853005A (en) | Immunohistochemical membrane staining section diagnosis method and device | |
| CN111524137A (en) | Cell identification counting method and device based on image identification and computer equipment | |
| CN108305253B (en) | Pathological image classification method based on multiple-time rate deep learning | |
| US9824189B2 (en) | Image processing apparatus, image processing method, image display system, and storage medium | |
| CN113962976B (en) | Quality evaluation method for pathological slide digital image | |
| CN111462076A (en) | Method and system for detecting fuzzy area of full-slice digital pathological image | |
| CN113344958B (en) | Microscopic imaging scanning method and scanning system | |
| CN113362331A (en) | Image segmentation method and device, electronic equipment and computer storage medium | |
| CN109035170A (en) | Adaptive wide-angle image correction method and device based on single grid chart subsection compression | |
| JP2015108837A (en) | Image processing apparatus and image processing method | |
| CN110837809A (en) | Blood automatic analysis method, blood automatic analysis system, blood cell analyzer, and storage medium | |
| CN111179173B (en) | Image splicing method based on discrete wavelet transform and gradient fusion algorithm | |
| CN113436206B (en) | Pathological tissue section scanning area positioning method based on cluster segmentation | |
| CN112464802B (en) | Automatic identification method and device for slide sample information and computer equipment | |
| CN109584368B (en) | Method and device for constructing three-dimensional structure of biological sample | |
| CN112580382B (en) | Two-dimensional code positioning method based on target detection | |
| CN117161004A (en) | Laser cleaning path planning method and system | |
| CN113052166A (en) | Pathological image display method and device | |
| CN115115919B (en) | Power grid equipment thermal defect identification method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: Room 6-1-604, No.55 Fuye street, Luotuo street, Zhenhai District, Ningbo City, Zhejiang Province Patentee after: Lansi (Ningbo) Intelligent Technology Co.,Ltd. Address before: Room 6-1-604, No.55 Fuye street, Luotuo street, Zhenhai District, Ningbo City, Zhejiang Province Patentee before: Ningbo Lanxi Biotechnology Co.,Ltd. |
|
| CP01 | Change in the name or title of a patent holder | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20210316 Assignee: Changsha Lansi Intelligent Technology Co.,Ltd. Assignor: Lansi (Ningbo) Intelligent Technology Co.,Ltd. Contract record no.: X2023980033656 Denomination of invention: A method and system for adaptive scanning control of pathological sections with dual objective lenses Granted publication date: 20210504 License type: Common License Record date: 20230321 |
|
| EE01 | Entry into force of recordation of patent licensing contract |