CN114202510B - Intelligent analysis system for pathological section image under microscope - Google Patents

Intelligent analysis system for pathological section image under microscope Download PDF

Info

Publication number
CN114202510B
CN114202510B CN202111332278.5A CN202111332278A CN114202510B CN 114202510 B CN114202510 B CN 114202510B CN 202111332278 A CN202111332278 A CN 202111332278A CN 114202510 B CN114202510 B CN 114202510B
Authority
CN
China
Prior art keywords
pathological section
image
microscope
pathological
model
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
Application number
CN202111332278.5A
Other languages
Chinese (zh)
Other versions
CN114202510A (en
Inventor
崔磊
刘建业
李涵生
董阳
王亚栋
杨林
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
NORTHWEST UNIVERSITY
Original Assignee
NORTHWEST UNIVERSITY
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 NORTHWEST UNIVERSITY filed Critical NORTHWEST UNIVERSITY
Priority to CN202111332278.5A priority Critical patent/CN114202510B/en
Publication of CN114202510A publication Critical patent/CN114202510A/en
Application granted granted Critical
Publication of CN114202510B publication Critical patent/CN114202510B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/217—Validation; Performance evaluation; Active pattern learning techniques
    • G06F18/2178—Validation; Performance evaluation; Active pattern learning techniques based on feedback of a supervisor
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H15/00—ICT specially adapted for medical reports, e.g. generation or transmission thereof
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/60—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices
    • G16H40/67—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the operation of medical equipment or devices for remote operation
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30024—Cell structures in vitro; Tissue sections in vitro
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30096—Tumor; Lesion
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30168—Image quality inspection

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • Public Health (AREA)
  • Epidemiology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Primary Health Care (AREA)
  • Evolutionary Biology (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Pathology (AREA)
  • Immunology (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Radiology & Medical Imaging (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Business, Economics & Management (AREA)
  • Business, Economics & Management (AREA)
  • Quality & Reliability (AREA)
  • Image Analysis (AREA)

Abstract

The invention relates to an intelligent analysis system for pathological section images under a microscope, which comprises a microscope, a microscope camera, a mechanical arm, a processor, a memory and a display; the microscope camera is used for collecting images of pathological sections under a microscope, and the mechanical arm is used for realizing section movement under the microscope objective. The processor receives pathological section images from the microscope camera, the system matches the digital pathological section images with an adaptive algorithm model, carries out algorithm processing on the pathological section images through intelligent diagnosis to obtain a preliminary diagnosis result, and outputs marked pathological section images; manually judging the pathological section image with the label, and if the labeling result meets the requirement, not correcting and labeling; otherwise, manually correcting the marking information on the pathological section image, and outputting a manual judgment result; and processing the manually corrected pathological section image, retraining the model through the local data, enhancing the suitability of the model to the local data, and outputting a final pathological report.

Description

Intelligent analysis system for pathological section image under microscope
Technical Field
The invention belongs to the technical field of pathological section diagnosis, and particularly relates to an intelligent analysis system for pathological section images under a microscope.
Background
At present, the medical department can be established by three-level hospitals and two-level comprehensive, chinese and western medicine combination, foot tumor, children and the patients with conditions in the gynaecology and obstetrics department hospitals.
The traditional microscope has high efficiency, accuracy, usability and cost performance to be improved, doctors need to switch between the screen and the microscope repeatedly, the judging effect is different from person to person, and pathological sections are identified manually by the doctors. If the visual field under the microscope is digitized, the accuracy and the real-time performance can meet the practical application requirements of pathological diagnosis, and the diagnosis conclusion can be better obtained by assisting a doctor, so that the reading pressure of the doctor is obviously relieved. Just like driving to step on the accelerator, the pathologist only needs to lightly step on the pedal, and the result can be accurately presented on the screen soon, so that accurate diagnosis and accurate treatment are realized.
The digital auxiliary film reading process of the deep learning is integrated and the traditional microscope is built, so that pathology research can obtain practical value on the windward vehicle in the current medical field, and an intelligent algorithm based on the deep learning is introduced into a place where feet can be found.
In China, the intelligent microscope commonly developed by gold domain medicine, tengxuan AI Lab and Shuyu optical technology has acquired registration certificate issued by NMPA, and is an intelligent microscope product which is approved for clinic in China for the first time. The intelligent microscope integrates the latest technology in the aspect of current pathological analysis and diagnosis, carries out multiple product iterations aiming at the workflow and habit of pathologists, and supports interpretation of the quantitative analysis scenes of common nuclear staining and membrane staining such as breast cancer Immunohistochemistry (IHC), ki67 (tumor cell proliferation index), ER (estrogen receptor), PR (progestin receptor), her2 (cell surface growth factor) and the like. The following defects still exist in the practical application process:
1. the intelligent microscope developed by Tencentrated corporation with other companies cannot remotely move the section observation position and the intelligent analysis time is long. Although the algorithm is rich and the automation degree is high, the developed integrated equipment can use the complete functions only after purchasing the complete set of software and hardware. Because most of intelligent microscopes integrating multiple functions are integrated equipment, the intelligent microscopes are high in selling price, and artificial intelligent algorithms developed by other people cannot be added independently according to actual disease diagnosis conditions of hospitals.
2. Because the current artificial intelligence algorithm processes the pathological image, a doctor can only passively accept the detection result and cannot manually correct the processing result. So doctors often challenge this black box mechanism that cannot be manually engaged.
3. Because of the differences between the microscopes and the staining agents in hospitals, most intelligent microscopes have poor diagnosis effect under a fixed model, poor adaptability to local mode processing and low accuracy of detection results.
Disclosure of Invention
Aiming at the defects or shortcomings of the prior art, the invention aims to provide an intelligent analysis system for pathological section images under a microscope.
In order to achieve the above task, the present invention adopts the following technical solutions:
an intelligent analysis system for pathological section images under a microscope is characterized by comprising:
a microscope for observing a pathological section;
the microscope camera is provided with an acquisition switch and is connected with a lens of the microscope and used for acquiring pathological section images under the microscope;
the mechanical arm is used for connecting a microscope moving hand wheel, controlling the longitudinal and horizontal movement of the moving platform to move the slice, and realizing the switching of pathological slice images under the microscope;
the processor is used for being in communication connection with the microscope camera, receiving pathological section images from the microscope camera, preprocessing the pathological section images, and carrying out cell detection and cell classification on the processed pathological section images;
one or more memories connected to the processor for storing programs and data;
the display is connected with the processor and used for receiving and displaying the marked pathological section image from the processor in real time;
the Bluetooth device is used for carrying out information interaction between the mechanical arm and the processor; and
one or more programs stored in the one or more memories and invoked for execution by the processor;
the program includes instructions for performing the steps of:
1) Acquiring digital pathological section images by a microscope camera;
2) Matching the digital pathological section image with an adaptive algorithm model, performing algorithm processing on the pathological section image through intelligent diagnosis to obtain a preliminary diagnosis result, and outputting the marked pathological section image;
3) Manually judging the pathological section image with the label, and if the labeling result meets the requirement, ensuring high accuracy and not carrying out correction labeling; otherwise, manually correcting the marking information on the pathological section image, and outputting a manual judgment result;
4) Processing the manually corrected pathological section image, and retraining the model through the local data, so as to enhance the suitability of the model to the local data;
5) And outputting a final pathology report.
According to the invention, the microscope camera is remotely operated to acquire digital pathological section images, and the pathological section is moved without manual operation, and the method specifically comprises the following steps:
judging whether the mechanical arm needs to be operated or not;
if the mechanical arm does not need to be moved, transmitting the currently acquired pathological section image to a processor, and processing and labeling the pathological section image; otherwise, the mechanical arm is moved, and the movement direction of the mechanical arm is judged;
the movement direction signal of the mechanical arm is transmitted through Bluetooth equipment, and a motor in the mechanical arm controls the mechanical arm to rotate so as to drive a microscope moving hand wheel to rotate;
and acquiring a new pathological section image, transmitting the new pathological section image to a processor, and carrying out algorithm processing on the pathological section image.
Further, the one or more programs include image blur recognition, interactive correction pathological section labeling processing, and retraining model optimization methods, such that the processing model is adaptive to local data; wherein:
the image fuzzy recognition is used for intelligently recognizing the micro deformation of the pathological section image under the microscope, and intelligently recognizing the movement, acquisition and selection algorithm processing of the pathological section image;
the interactive correction pathological section labeling process is used for labeling information of pathological section images processed by a doctor interactive correction algorithm, and can add, delete and cancel correction operations on the pathological section image labels;
the retraining model optimization method is used for retraining all the corrected pathological section images so that the corrected pathological section image model is adapted to local data and the adaptability of the corrected pathological section images is improved;
the program receives the pathological section image, carries out fuzzy recognition detection, judges whether the current pathological section image moves or not, releases the collected pathological section image if the current pathological section image does not move, and does not carry out algorithm processing;
if the images are the pathological section images acquired after the movement, selecting an algorithm to process the pathological section images and outputting pathological section image information with labels;
after receiving the pathological section image, displaying the pathological section image with the label, and judging whether the label needs to be corrected by a doctor;
if correction is needed, the doctor performs the labeling operation of adding and deleting the selected area, stores the corrected labeled pathological section image, performs model training on the corrected labeled pathological section image through a retraining model optimization method, and improves the adaptability of the corrected pathological section image;
carrying out manual correction confirmation on the correction result;
displaying the image information of the pathological section with the label, and performing manual judgment;
if the image annotation information is judged to be complete and correct manually, outputting a pathological image report as a final report;
otherwise, correcting the image marking information, and manually performing operations such as adding, deleting, withdrawing and the like on the selected area, and correcting the marking information of the selected area to ensure that the image meets the actual diagnosis requirement;
outputting a pathological image report and storing labels and image information;
and if model optimization is not required to be trained, directly outputting a final pathological section image report.
Preferably, the image blur recognition method specifically includes:
and calculating image edge information by using a Laplacian operator to obtain the image blurring degree and scoring the image blurring degree, performing sliding filtering on the blurring degree score to obtain a final score, and performing intelligent image analysis if the final score is higher than a threshold value, otherwise, performing no analysis.
Further preferably, the retraining model optimization retrains the model through local correction data, so as to improve the adaptability of the model to the local data, and specifically includes:
firstly, a doctor starts a trained model to diagnose a pathological section transmitted by a microscope camera, and different labeling information is marked on different cells to distinguish the cells;
secondly, because the trained model uses less local data, the model is likely to have lower suitability for the local data, and doctors can modify, replace, add, delete and the like labels;
then adding the modified labeling information into the original pathological section image training set, further obtaining local data, and finally improving the problem of data inconsistency;
finally, when the number of pathological section images modified by the doctor reaches a certain number, the doctor can select a retraining model optimization mode to optimize the model, and the model after optimization is more adaptive to local data, so that the accuracy of intelligent analysis is improved.
The intelligent analysis system for pathological section images under a microscope has the following beneficial effects:
(1) Compared with the prior intelligent microscope equipment, the system integrated operation microscope is realized, so that the doctor is prevented from frequently switching operation between the computer and the microscope. Compared with the traditional microscope equipment, the intelligent analysis system for the pathological section images under the microscope has the advantages that a doctor needs to frequently and manually adjust the movable hand wheel, and the doctor can directly operate the movable hand wheel to realize the switching of the observation field of view through the intelligent analysis system for the pathological section images under the microscope, so that the operability is greatly improved.
(2) Compared with the prior system, the system has the remarkable advantages of higher real-time performance and greatly reduces the time delay required by the system to display the detection result. Compared with the prior device, the real-time display is low because the user needs to run the detection program for 2s-3s after clicking the detection button. According to the invention, a fuzzy recognition algorithm is adopted, and the image information is automatically detected and calculated after the object lens stops moving, and the detection is performed by using the time period that the doctor stops moving the lens and clicks the detection button, so that the subjective use feeling of the doctor is improved, and the real-time performance of the system is improved.
(3) Compared with the traditional intelligent auxiliary diagnosis system, the diagnosis system has the remarkable advantage that a doctor can manually correct a pathological detection result during diagnosis. The doctor can manually insert, delete, cancel and the like the detection conclusion, so that the problems that the doctor can only passively accept a system detection report under the traditional intelligent auxiliary diagnosis system, the analysis result cannot be corrected and the manual operation is repeated are solved. According to the invention, the training model can be continuously corrected according to the manual correction data, so that doctors can further participate in the diagnosis process, and the manual operation repetition rate is reduced.
(4) The problem that the data of the model training data set and the data used by the hospital are inconsistent is solved by using the data of the retraining training hospital, so that the generalization capability of the model is greatly improved, and the accuracy of system judgment is further improved.
Drawings
FIG. 1 is a block diagram of an intelligent analysis system for images of pathological sections under a microscope according to the present invention;
FIG. 2 is a flow chart of the operation of the intelligent analysis system for images of pathological sections under a microscope of the present invention;
FIG. 3 is a flow chart of intelligent analysis of pathological section images;
FIG. 4 is a schematic illustration of a robotic arm interaction flow;
FIG. 5 is a schematic diagram of a blur detection flow;
FIG. 6 is a flow chart of medical record preservation by the intelligent analysis system for pathological section images under a microscope of the invention;
FIG. 7 is an interface design of the intelligent analysis system for pathological section images under a microscope of the invention;
FIG. 8 is an illustration of the processing operations performed by the intelligent analysis system for microscopic pathological section images according to the present invention in TCT mode interaction;
FIG. 9 is a diagram showing the insertion operation of the intelligent analysis system for pathological section images under a microscope according to the present invention in TCT mode interaction;
FIG. 10 is a diagram showing the deletion operation of the intelligent analysis system for pathological section images under a microscope according to the present invention in TCT mode interaction;
FIG. 11 is a diagram showing the insertion operation of the intelligent analysis system for pathological section images under a microscope according to the present invention in the interaction of the ki67 mode;
FIG. 12 is a diagram showing the deletion operation of the intelligent analysis system for microscopic pathological section images according to the present invention in the interaction of the ki67 mode;
FIG. 13 is a diagram showing the operation of the intelligent analysis system for microscopic pathological section images according to the present invention in the interaction of the ki67 mode;
FIG. 14 is an illustration of the operation of the intelligent analysis system for microscopic pathological section images under the interaction of the training system of the present invention.
The invention is described in further detail below with reference to the drawings and examples.
Detailed Description
The design concept of the invention is that firstly, the problem that the existing intelligent microscope cannot remotely move the slice observation position is solved, and the slice position is changed by controlling the moving hand wheel; and immediately processing the acquired pathological section images by a fuzzy recognition method after the microscope stops moving, so that the analysis result can be displayed in real time.
According to the problem that the processing result cannot be corrected manually in the existing problems, the problem that the adaptability of the local model is poor is solved, the retraining model-based optimization method is provided, the processing model is optimized continuously, and the analysis accuracy of the system is improved.
Referring to fig. 1, this embodiment provides an intelligent analysis system for pathological section images under a microscope, including:
a microscope for observing a pathological section;
the microscope camera is provided with an acquisition switch and is connected with a lens of the microscope and used for acquiring pathological section images under the microscope;
the mechanical arm is used for connecting a microscope moving hand wheel, controlling the longitudinal and horizontal movement of the moving platform to move the slice, and realizing the switching of pathological slice images under the microscope;
the processor is used for being in communication connection with the microscope camera, receiving pathological section images from the microscope camera, preprocessing the pathological section images, and carrying out cell detection and cell classification on the processed pathological section images;
one or more memories connected to the processor for storing programs and data;
the display is connected with the processor and used for receiving and displaying the marked pathological section image from the processor in real time;
the Bluetooth device is used for carrying out information interaction between the mechanical arm and the processor; and
one or more programs stored in the one or more memories and invoked for execution by the processor;
the program includes instructions for performing the steps of:
1) Acquiring digital pathological section images by a microscope camera;
2) Matching the digital pathological section image with an adaptive algorithm model, performing algorithm processing on the pathological section image through intelligent diagnosis to obtain a preliminary diagnosis result, and outputting the marked pathological section image;
3) Manually judging the pathological section image with the label, and if the labeling result meets the requirement, ensuring high accuracy and not carrying out correction labeling; otherwise, manually correcting the marking information on the pathological section image, and outputting a manual judgment result;
4) Processing the manually corrected pathological section image, and retraining the model through the local data, so as to enhance the suitability of the model to the local data;
5) And outputting a final pathology report.
When the intelligent analysis system for pathological section images under the microscope is used, the acquisition switch is turned on, and the microscope camera acquires pathological section images of the cell smear under the microscope and sends the pathological section images to the processor; the processor performs cell detection and cell classification on the pathological section image from the microscope camera, marks the position of the cell, the cell type and the corresponding confidence coefficient on the pathological section image, and outputs the marked pathological section image; the display receives and displays the marked pathological section image from the processor; the Bluetooth equipment is connected with the processor and the mechanical arm, and sends a moving signal to the mechanical arm to control the mechanical arm to perform forward rotation and reverse rotation, so that the switching of the moving angle and the moving direction is realized; the mechanical arm controls the microscope to move the hand wheel through the up-down, left-right direction keys in the system to realize the horizontal or longitudinal movement of the moving platform, thereby changing the pathological section position observed by the microscope and switching the local position of the pathological section in real time for observation.
The mechanical arm comprises the following functions:
firstly, the Bluetooth equipment is connected with a processor and a mechanical arm, so that normal communication of both sides is ensured, and the operation of a microscope moving hand wheel is realized;
secondly, the system detects the current moving operation direction of the user, and transmits signals to the mechanical arm through the Bluetooth device;
and finally, a motor of the mechanical arm operates the mechanical arm to rotate a microscope moving hand wheel according to the transmission signal, so that the observation area is switched.
In this embodiment, microscope camera gathers digital pathological section image and adopts remote operation, need not artifical manual operation and removes pathological section, specifically includes:
judging whether the mechanical arm needs to be operated or not;
if the mechanical arm does not need to be moved, transmitting the currently acquired pathological section image to a processor, and processing and labeling the pathological section image; otherwise, the mechanical arm is moved, and the movement direction of the mechanical arm is judged;
the movement direction signal of the mechanical arm is transmitted through Bluetooth equipment, and a motor in the mechanical arm controls the mechanical arm to rotate so as to drive a microscope moving hand wheel to rotate;
and acquiring a new pathological section image, transmitting the new pathological section image to a processor, and carrying out algorithm processing on the pathological section image.
The one or more programs comprise image fuzzy recognition, interactive correction pathological section labeling processing and retraining model optimization methods, so that the processing model is adaptive to local data; wherein:
the image fuzzy recognition is used for intelligently recognizing the micro deformation of the pathological section image under the microscope, and intelligently recognizing the movement, acquisition and selection algorithm processing of the pathological section image;
the interactive correction pathological section labeling process is used for labeling information of pathological section images processed by a doctor interactive correction algorithm, and can add, delete and cancel correction operations on the pathological section image labels;
the retraining model optimization method is used for retraining all the corrected pathological section images so that the corrected pathological section image model is adapted to local data and the adaptability of the corrected pathological section images is improved;
the program receives the pathological section image, carries out fuzzy recognition detection, judges whether the current pathological section image moves or not, releases the collected pathological section image if the current pathological section image does not move, and does not carry out algorithm processing;
if the images are the pathological section images acquired after the movement, selecting an algorithm to process the pathological section images and outputting pathological section image information with labels;
after receiving the pathological section image, displaying the pathological section image with the label, and judging whether the label needs to be corrected by a doctor;
if correction is needed, the doctor performs the labeling operation of adding and deleting the selected area, stores the corrected labeled pathological section image, performs model training on the corrected labeled pathological section image through a retraining model optimization method, and improves the adaptability of the corrected pathological section image;
carrying out manual correction confirmation on the correction result;
displaying the image information of the pathological section with the label, and performing manual judgment;
if the image annotation information is judged to be complete and correct manually, outputting a pathological image report as a final report;
otherwise, correcting the image marking information, and manually performing operations such as adding, deleting, withdrawing and the like on the selected area, and correcting the marking information of the selected area to ensure that the image meets the actual diagnosis requirement;
outputting a pathological image report and storing labels and image information;
and if model optimization is not required to be trained, directly outputting a final pathological section image report.
In this embodiment, the retraining model optimization retrains the model through local correction data, so as to improve adaptability of the model to the local data, and specifically includes:
firstly, a doctor starts a trained model to diagnose a pathological section transmitted by a microscope camera, and different labeling information is marked on different cells to distinguish the cells;
secondly, because the trained model uses less local data, the model is likely to have lower suitability for the local data, and doctors can modify, replace, add, delete and the like labels;
then adding the modified labeling information into the original pathological section image training set, further obtaining local data, and finally improving the problem of data inconsistency;
finally, when the number of pathological section images modified by the doctor reaches a certain number, the doctor can select to click the upper left corner retraining button to optimize the model, and the model after optimization is more adaptive to local data, so that the accuracy of intelligent analysis is improved.
Referring to fig. 2, the operation procedure of the intelligent analysis system for pathological section image under microscope of the present embodiment is as follows:
for the local mode of operation:
uploading pathological sections, selecting a pathological section image processing mode, carrying out algorithm processing on the pathological sections, clicking a doctor, namely viewing and displaying a processing result, setting various programs of clicking insertion, clicking deletion, clicking withdrawal, clicking storage and clicking retraining, and manually adding processing information in clicking insertion; in clicking "delete", deleting unnecessary processing results; in clicking "undo", the previous operation is resumed; in clicking "save", save the current operation result and patient information, etc.; in click "retraining", a model is built that fits the local data by retraining the local data set.
For the real-time mode of operation:
and receiving pathological section data from a microscope camera, when an acquisition switch is turned on, acquiring current pathological section data, and performing fuzzy detection on whether the camera moves, if so, displaying pathological section images, then selecting a pathological section image processing mode, performing algorithm processing on the pathological section, and performing the following operation in the same local operation mode.
The intelligent analysis system for pathological section images under the microscope can process two different pathological section images, one is used for automatically judging ki67 indexes, can be used for judging the malignancy degree of tumors, and the other is used for detecting TCT section positive cells of cervical cancer and can be used for screening early cervical cancer. The intelligent analysis system marks the position of the cell, the cell type and the corresponding confidence coefficient on the pathological image and outputs the marked pathological image; and the display is connected with the processor and is used for receiving and displaying the marked pathological section image. The acquisition switch is arranged on the microscope camera and controls the operation of the microscope camera.
The intelligent analysis of the pathological section image is shown in fig. 3, firstly, judging whether the current operation mode is a real-time operation mode or a local operation mode, if the current operation mode is the local operation mode, directly acquiring pathological section data, then selecting a processing mode of ki67 or TCT, carrying out algorithm processing on the pathological section data image, manually correcting an algorithm result by carrying out operations of adding, deleting, withdrawing, storing and the like on the pathological section data image, writing and storing medical record information, and retraining the stored local data to enable the model to adapt to the local data.
If the operation mode is the real-time operation mode, the fuzzy recognition is carried out on the pathological section area, the micro deformation of the pathological section image under the microscope is reduced, whether the current pathological section moves or not is judged, then the movement of the pathological section is controlled by a mobile hand wheel of the microscope through a mechanical arm control, the image information of the pathological section area needing to be processed is obtained, and the operation is the same as the local operation mode.
The interaction flow of the mechanical arm is shown in fig. 4, the mechanical arm is connected with the processor through the Bluetooth device, the user writes the mobile button, the Bluetooth device transmits signals to the mechanical arm, and the mechanical arm rotates and moves the microscope hand wheel according to the signals.
The fuzzy recognition detection flow is shown in fig. 5, and comprises the following steps:
(1) Performing edge detection on the current video stream image data by using a Laplacian operator to obtain the blurring degree of the video stream image data, and scoring to obtain curr_var;
(2) Performing sliding filtering on the fuzzy degree score curr_var to obtain a comprehensive fuzzy degree score image_var=0.5×last_var+0.5×curr_var;
(3) And performing intelligent detection on the video stream image data with the comprehensive blurring degree score higher than a threshold value, and performing no detection when the comprehensive blurring degree score is lower than the threshold value.
The medical record preservation flow chart is shown in fig. 6, and comprises the following steps:
(1) Establishing basic information containing patients, including names, sexes, ages, clinic numbers, hospitalization numbers, bed numbers, pathology numbers, delivery hospitals, delivery departments, delivery doctors, delivery dates, material taking parts, clinical diagnoses and naked eyes, wherein the information of the clinical diagnoses can be written in by the background and can be manually modified by doctors, and other information can be manually written in by the doctors;
(2) After clicking the save button, the system saves the patient medical record, the original slice image, the processed slice image and the processed image marking information to a folder;
(3) Submitting medical records and detection results;
the interface design of the intelligent analysis system for the pathological section image under the microscope is shown in fig. 7, and the interface design comprises the following contents:
(1) Tool bars, including menu buttons File, edit, view, window, help and Retrain, tool buttons "open file", "open folder", "save", "print" and "help". The toolbar is used for performing system basic configuration operation;
(2) An operation mode selection field including a mode selection 1 selecting an algorithm processing mode "TCT processing" or "ki67 processing", and a mode selection 2 selecting a "local operation" or "real-time operation" mode;
(3) Interactive toolbar including "operate", "draw" and "save" interactive buttons. The operational functions include diagnostic, insert, delete and undo functions. The drawing function includes drawing labeling information of the type including a dot, a rectangular frame, and a free pen. The preservation function stores the pathological section image corrected by the interactive operation to a local appointed path;
(4) Patient medical record information including name, sex, age, clinic number, hospitalization number, bed number, pathology number, hospital for censoring, department for censoring, doctor for censoring, date for censoring, material taking location, clinical diagnosis, macroscopic, wherein information of clinical diagnosis can be written in by the background and doctor can also be manually modified, and other information can be written in by doctor manually. After clicking the save button, the system saves the patient medical record, the original slice image, the processed slice image and the processed image annotation information in a folder.
The intelligent analysis system for pathological section images under the microscope of the embodiment has the following functions:
(1) The method comprises the steps of selecting a pathological section image processing mode after a microscope is started, designing a fuzzy recognition algorithm aiming at pathological section image analysis, immediately processing the pathological section image after detecting that the pathological section image stops moving, and displaying an analysis result in real time.
(2) After clicking the "diagnosis" button, the processing results are displayed entirely, or the processing results are displayed by selecting the custom area through the "drawing" button.
(3) In the TCT mode, clicking an 'insert' button, clicking a 'draw' button, selecting a disease type identifier, and drawing a mark frame and the disease type identifier according to the sliding range of a mouse on a pathological section image; after clicking the 'delete' function, deleting a mark frame containing the coordinates on the pathological picture according to the mouse click coordinates and a disease type mark; after clicking the cancel function, recovering the deleted mark frame and disease identification in the rectangular selected area according to the sliding range of the mouse; after clicking the "save" button, three data are saved:
(1) processing results by the current system;
(2) cutting the original pathological picture according to the mark frame to obtain a local picture;
(3) TXT files containing all the logo boxes and disease identification information in the local pictures.
(4) In the ki67 mode, click on the "insert" function, and then click on the "draw" button to select the insert mode as "rectangular box" or "free pen". Selecting an insertion area according to the corresponding insertion mode, and inserting the mark point into the area; clicking the "delete" function and then clicking the "draw" button selects the delete mode as "point", "rectangular box" or "free pen". Deleting all the mark points in the deleting area selected according to the corresponding deleting mode; clicking the "cancel" function, and clicking the "draw" button selects the cancel mode as "rectangular box" or "free pen". Restoring and displaying all deleted mark points in the area according to the revocation area selected by the corresponding revocation mode; clicking the save function to save the current processing result to the designated address.
1) After the pathological section image is processed and manually corrected, medical records containing basic information of a patient, information of detected pathological sections and detection results are established and a detection report is submitted. The patient medical record comprises name, sex, age, clinic number, hospitalization number, bed number, pathology number, examination room, examination department, examination doctor, examination date, material taking part, clinical diagnosis, and naked eye, wherein the information of clinical diagnosis can be written in by the background and can be manually modified by the doctor, and other information can be manually written in by the doctor. After clicking the save button, the system saves the patient medical record, the original slice image, the processed slice image and the processed image annotation information in a folder.
2) The retraining operation is performed following the steps of: firstly, a doctor starts a trained model to diagnose pathological section images transmitted by a microscope camera, and different labeling information is marked on different cells to distinguish the cells. Secondly, because the trained model uses less local data, the model may not have a very good effect on the local data, so that the accuracy of model labeling needs to be judged, and a doctor can modify, replace, add, delete and the like the labeling. And then adding the modified annotation information into the original training set, so that local data can be obtained, and the problem of data inconsistency is further solved. Finally, when the number of the images modified by the doctor reaches a certain number, the doctor can select to click the upper left corner retraining button to optimize the model, the model after optimization is more suitable for local data, and better effects can be obtained in the aspects of accuracy and the like.
The operation interaction mode comprises the following contents:
(1) The insertion method of the present embodiment differs according to the processing mode.
1) In the TCT mode, firstly, a disease type mark is selected, the corresponding color of the corresponding mark is obtained, and secondly, a rectangular area formed by a starting point coordinate (x 0, y 0) clicked by a mouse and a released ending point coordinate (x 1, y 1) is used as a mark frame. The area surrounded by the starting point of the marker frame and the lower right (x 1-x 0)/10, y0-abs (y 1-y 0)/10) is used as a disease type identification display area.
2) The drawing mode is first selected to be "rectangular box" or "free pen" in the ki67 mode. In the "rectangular frame" mode, it is determined whether there is a rectangular region in which the dot coordinates are located according to the rectangular region formed by the start point coordinates (x 0, y 0) and the released end point coordinates (x 1, y 1), and the dot is inserted into the rectangular region enclosed. In the free pen mode, a point coordinate set of a closed graph is formed by a mouse sliding track, whether the point coordinate is located in the closed graph is judged, and the existing point coordinate is inserted into a pathological image.
(2) The deletion method in this embodiment differs according to the processing mode.
1) After clicking the 'delete' button in the TCT mode, clicking a mark frame to be deleted in the pathological graph, judging the mark frame containing the click coordinates, and deleting.
2) After clicking a 'delete' button in the Ki67 mode, selecting a 'point' deleting mode, clicking a point to be deleted, and deleting the point information according to click coordinates; selecting a 'rectangular frame' deleting mode, namely deleting a rectangular area formed by starting point coordinates (x 0, y 0) clicked by a mouse and released ending point coordinates (x 1, y 1), and deleting point coordinate information contained in the area; and selecting a free pen deleting mode, forming a point coordinate set of a closed graph by a mouse sliding track, judging that the point set encloses points contained in a closed area, and deleting point coordinate information.
(3) The revocation method of the present embodiment differs according to the processing mode:
1) After clicking the cancel button in the TCT mode, the deleted point coordinate information of the rectangular area formed by the start point coordinate (x 0, y 0) clicked by the mouse and the released end point coordinate (x 1, y 1) is restored and displayed.
2) After clicking a cancel button in a ki67 mode, selecting a rectangular frame cancel mode, and recovering the deleted point coordinate information of the rectangular region formed by the start point coordinates (x 0, y 0) clicked by a mouse and the released end point coordinates (x 1, y 1); and selecting a free pen withdrawal mode, forming a point coordinate set of a closed graph by a mouse sliding track, and recovering the point coordinate set to form deleted point coordinate information contained in a closed area.
(4) The storage method of the present embodiment differs according to the processing mode.
1) After clicking the "save" button in TCT mode:
firstly, for the preservation of a detection result, firstly, acquiring a pixmap displayed in a current window, storing the pixmap as a temPix, drawing all mark frames and disease type identifications in display on the temPix, secondly, converting the temPix into an array form, setting the picture size to be 1920 x 1080, and then, preserving the detection result to a specified path;
secondly, intercepting and storing the area with the mark frame, firstly acquiring the picture array before processing, secondly intercepting the area with the size of 512 x 512 according to the position information of the mark frame, and reducing the area to the size of 256 x 256. Then, the original coordinates of the intercepted area are stored, whether the mark frame is completely contained in the intercepted area or not is judged before interception, interception is not carried out any more if the screenshot of the area exists, and finally, the intercepted picture is stored to \model\yolov5\TCT\images\train;
thirdly, for storage of the cut-out region mark frame information and the disease type identification information, after each cut-out region, the information is written into TXT files with the same name as the corresponding cut-out region according to the class (disease type information), x=x_center/width (picture width), y=y_center/height (picture height), w= (x_max-x_min)/width, h= (y_max-y_min)/height, and finally the files are stored to the same name.
2) After clicking a 'save' button in the Ki67 mode, firstly, selecting a save path of a processing result by a user, secondly, acquiring pixmap of a current window, converting the pixmap into an array format, setting the picture size to 1920 x 1080, and finally, writing the displayed mark point information into the path designated by the user through OpenCV.
The drawing interaction mode comprises the following steps:
(1) After the 'point' interaction mode is selected, judging whether a mark point exists in a 2 x 2 pixel range by taking the coordinate as the center or not by acquiring the coordinate of the clicking position of the mouse.
(2) After the 'rectangular frame' interaction mode is selected, the starting point coordinates (x 0, y 0) and the released ending point coordinates (x 1, y 1) of the mouse click are obtained, the starting point coordinates (x 0, y 0) are used as the left top vertex of the rectangular area, and x1-x0 and y1-y0 are used as the width and height of the rectangular area respectively to draw the rectangular area. For the coordinates (x, y) in turn, point coordinate information satisfying the condition of x0< =x < =x1 and y0< =y < =y1 is output.
(3) After the 'free pen' interaction mode is selected, the coordinates of the selected points to be judged and the track information in the list are judged by acquiring the sliding track of the mouse and storing the track as the list. Firstly, whether the point is on the track or not is judged, and secondly, whether the point (px, py) is positioned in the inner part of the track surrounding the graph or not is judged. If the ordinate is between any two points (sx, sy) and (tx, ty) on the track, the relation between the abscissa px and the (tx-sx)/(ty-sy) coordinate of x=sx+ (py-sy) is determined. If x=px, the point (px, py) is in the irregular area, and if x > px, the relationship between the point (px, py) and any other two points on the track is continued to be determined. Finally, outputting point information on the track or in the enclosed area.
In addition, the intelligent analysis system for the pathological section image under the microscope provided by the embodiment also carries out the following design on the following interactive operation interface:
FIG. 8 is a diagram showing the processing operation under TCT interaction, showing the display form after inserting the flag information under TCT; FIG. 9 is a diagram showing an insert operation performed in TCT mode interaction, enabling a physician to interactively perform a logo information insertion; FIG. 10 shows a deletion operation demonstration diagram under TCT mode interaction, so that a doctor can delete and correct misjudged labeling information; FIG. 11 is a diagram showing the insertion operation performed in the interaction of the ki67 mode, and the classification of cells in the detection area is noted; FIG. 12 is a diagram illustrating deletion operations performed in the ki67 mode interaction, with selected region annotation information being deleted; FIG. 13 is a diagram illustrating an undo operation in a ki67 mode interaction that undoes a modification operation to a selected region to facilitate modification of annotations; FIG. 14 shows an operational illustration under retraining system interactions to improve model accuracy by retraining an optimized model.
In summary, the intelligent analysis system for pathological section images under a microscope provided by the embodiment dynamically acquires pathological section image data by using a bit data compression technology, adopts a fuzzy recognition method to recognize moving pathological section images and reduce errors caused by micro deformation of the pathological section images, performs intelligent analysis under static pathological section images, and improves the real-time performance of the system. After the intelligent analysis result is obtained, a doctor can manually correct the detection result and participate in pathological section image report generation, so that the method accords with the actual workflow and improves the accuracy of the final result. Because of the problems of inconsistent data and the like caused by different microscopes or colorants used by different hospitals or scientific research institutions, the adaptability of the model and the local data is low, the system trains out the model suitable for the local data by using the acquired local data corrected by doctors and the method of labeling and retraining, finally improves the precision of the model, and realizes the optimization of algorithms.

Claims (3)

1. An intelligent analysis system for pathological section images under a microscope is characterized by comprising:
a microscope for observing a pathological section;
the microscope camera is provided with an acquisition switch and is connected with a lens of the microscope and used for acquiring pathological section images under the microscope;
the mechanical arm is used for connecting a microscope moving hand wheel, controlling the longitudinal and horizontal movement of the moving platform to move the slice, and realizing the switching of pathological slice images under the microscope;
the processor is used for being in communication connection with the microscope camera, receiving pathological section images from the microscope camera, preprocessing the pathological section images, and carrying out cell detection and cell classification on the processed pathological section images;
one or more memories connected to the processor for storing programs and data;
the display is connected with the processor and used for receiving and displaying the marked pathological section image from the processor in real time;
the Bluetooth device is used for carrying out information interaction between the mechanical arm and the processor; and
one or more programs stored in the one or more memories and invoked for execution by the processor;
the program includes instructions for performing the steps of:
1) Acquiring digital pathological section images by a microscope camera;
2) Matching the digital pathological section image with an adaptive algorithm model, performing algorithm processing on the pathological section image through intelligent diagnosis to obtain a preliminary diagnosis result, and outputting the marked pathological section image;
3) Manually judging the pathological section image with the label, and if the labeling result meets the requirement, ensuring high accuracy and not carrying out correction labeling; otherwise, manually correcting the marking information on the pathological section image, and outputting a manual judgment result;
4) Processing the manually corrected pathological section image, and retraining the model through the local data, so as to enhance the suitability of the model to the local data;
5) Outputting a final pathology report;
the one or more programs comprise image fuzzy recognition, interactive correction pathological section labeling processing and retraining model optimization methods, so that the processing model is adaptive to local data; wherein:
the image fuzzy recognition is used for intelligently recognizing the micro deformation of the pathological section image under the microscope, and intelligently recognizing the movement, acquisition and selection algorithm processing of the pathological section image;
the interactive correction pathological section labeling process is used for labeling information of pathological section images processed by a doctor interactive correction algorithm, and can add, delete and cancel correction operations on the pathological section image labels;
the retraining model optimization method is used for retraining all the corrected pathological section images so that the corrected pathological section image model is adapted to local data and the adaptability of the corrected pathological section images is improved;
the program receives the pathological section image, carries out fuzzy recognition detection, judges whether the current pathological section image moves or not, releases the collected pathological section image if the current pathological section image does not move, and does not carry out algorithm processing;
if the images are the pathological section images acquired after the movement, selecting an algorithm to process the pathological section images and outputting pathological section image information with labels;
after receiving the pathological section image, displaying the pathological section image with the label, and judging whether the label needs to be corrected by a doctor;
if correction is needed, the doctor performs the labeling operation of adding and deleting the selected area, stores the corrected labeled pathological section image, performs model training on the corrected labeled pathological section image through a retraining model optimization method, and improves the adaptability of the corrected pathological section image;
carrying out manual correction confirmation on the correction result;
displaying the image information of the pathological section with the label, and performing manual judgment;
if the image annotation information is judged to be complete and correct manually, outputting a pathological image report as a final report;
otherwise, correcting the image marking information, and manually performing adding, deleting and withdrawing operations on the selected area to correct the marking information of the selected area so that the image meets the actual diagnosis requirement;
outputting a pathological image report and storing labels and image information;
if model optimization is not required to be trained, directly outputting a final pathological section image report;
the image blurring recognition method specifically comprises the following steps:
and calculating image edge information by using a Laplacian operator to obtain the image blurring degree and scoring the image blurring degree, performing sliding filtering on the blurring degree score to obtain a final score, and performing intelligent image analysis if the final score is higher than a threshold value, otherwise, performing no analysis.
2. The intelligent analysis system for microscopic pathological section images according to claim 1, wherein the microscope camera is remotely operated to acquire digital pathological section images, and the pathological section is moved without manual operation, and the intelligent analysis system specifically comprises:
judging whether the mechanical arm needs to be operated or not;
if the mechanical arm does not need to be moved, transmitting the currently acquired pathological section image to a processor, and processing and labeling the pathological section image; otherwise, the mechanical arm is moved, and the movement direction of the mechanical arm is judged;
the movement direction signal of the mechanical arm is transmitted through Bluetooth equipment, and a motor in the mechanical arm controls the mechanical arm to rotate so as to drive a microscope moving hand wheel to rotate;
and acquiring a new pathological section image, transmitting the new pathological section image to a processor, and carrying out algorithm processing on the pathological section image.
3. The intelligent analysis system for microscopic pathological section images according to claim 1, wherein the retraining model optimization retrains the model by locally correcting data, and improves the adaptability of the model to the local data, and specifically comprises:
firstly, a doctor starts a trained model to diagnose a pathological section transmitted by a microscope camera, and different labeling information is marked on different cells to distinguish the cells;
secondly, because the trained model uses less local data, the model has low suitability for the local data, and doctors can modify, replace, add and subtract labels;
then adding the modified labeling information into the original pathological section image training set, further obtaining local data, and finally improving the problem of data inconsistency;
finally, when the number of pathological section images modified by doctors reaches a certain number, a model optimization mode can be selected to optimize the model, and the model after optimization is more adaptive to local data, so that the accuracy of intelligent analysis is improved.
CN202111332278.5A 2021-11-11 2021-11-11 Intelligent analysis system for pathological section image under microscope Active CN114202510B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111332278.5A CN114202510B (en) 2021-11-11 2021-11-11 Intelligent analysis system for pathological section image under microscope

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111332278.5A CN114202510B (en) 2021-11-11 2021-11-11 Intelligent analysis system for pathological section image under microscope

Publications (2)

Publication Number Publication Date
CN114202510A CN114202510A (en) 2022-03-18
CN114202510B true CN114202510B (en) 2024-01-19

Family

ID=80647355

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111332278.5A Active CN114202510B (en) 2021-11-11 2021-11-11 Intelligent analysis system for pathological section image under microscope

Country Status (1)

Country Link
CN (1) CN114202510B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116364229B (en) * 2023-04-20 2023-11-10 北京透彻未来科技有限公司 Intelligent visual pathological report system for cervical cancer anterior lesion coning specimen

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110763678A (en) * 2019-09-12 2020-02-07 杭州迪英加科技有限公司 Pathological section interpretation method and system
CN110763677A (en) * 2019-09-12 2020-02-07 杭州迪英加科技有限公司 Thyroid gland frozen section diagnosis method and system
WO2020038245A1 (en) * 2018-08-21 2020-02-27 腾讯科技（深圳）有限公司 Pathological microscope, display module, control method and device, and storage medium
CN112132772A (en) * 2019-06-24 2020-12-25 杭州迪英加科技有限公司 Pathological section real-time interpretation method, device and system
CN113130049A (en) * 2021-04-15 2021-07-16 黑龙江机智通智能科技有限公司 Intelligent pathological image diagnosis system based on cloud service
WO2021196632A1 (en) * 2020-03-30 2021-10-07 中国科学院深圳先进技术研究院 Intelligent analysis system and method for panoramic digital pathological image

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020038245A1 (en) * 2018-08-21 2020-02-27 腾讯科技（深圳）有限公司 Pathological microscope, display module, control method and device, and storage medium
CN112132772A (en) * 2019-06-24 2020-12-25 杭州迪英加科技有限公司 Pathological section real-time interpretation method, device and system
CN110763678A (en) * 2019-09-12 2020-02-07 杭州迪英加科技有限公司 Pathological section interpretation method and system
CN110763677A (en) * 2019-09-12 2020-02-07 杭州迪英加科技有限公司 Thyroid gland frozen section diagnosis method and system
WO2021196632A1 (en) * 2020-03-30 2021-10-07 中国科学院深圳先进技术研究院 Intelligent analysis system and method for panoramic digital pathological image
CN113130049A (en) * 2021-04-15 2021-07-16 黑龙江机智通智能科技有限公司 Intelligent pathological image diagnosis system based on cloud service

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
王莹 ; 段佳佳 ; 祝闯 ; 刘军 ; 董慧慧 ; 齐勇刚 ; 金木兰 ; .临床结直肠病理切片图像的自动辅助诊断.临床与实验病理学杂志.2018,(第10期),全文. *

Also Published As

Publication number Publication date
CN114202510A (en) 2022-03-18

Similar Documents

Publication Publication Date Title
CN109920518B (en) Medical image analysis method, medical image analysis device, computer equipment and storage medium
US7945083B2 (en) Method for supporting diagnostic workflow from a medical imaging apparatus
CN107296650A (en) Intelligent operation accessory system based on virtual reality and augmented reality
US5740267A (en) Radiographic image enhancement comparison and storage requirement reduction system
CN110473186B (en) Detection method based on medical image, model training method and device
US20130251233A1 (en) Method for creating a report from radiological images using electronic report templates
CN108921854B (en) Method and system for labeling irregular lesion region of gastrointestinal endoscope image
CN111430014B (en) Glandular medical image display method, glandular medical image interaction method and storage medium
WO2012105909A1 (en) An imaging system and method
US11344281B2 (en) Ultrasound visual protocols
CN112365438B (en) Pelvis parameter automatic measurement method based on target detection neural network
CN114202510B (en) Intelligent analysis system for pathological section image under microscope
CN113485555B (en) Medical image film reading method, electronic equipment and storage medium
CN104546006A (en) Accurately-positioning three-dimensional medical ultrasonic image workstation and system
CN116779093B (en) Method and device for generating medical image structured report and computer equipment
CN112489053B (en) Tongue image segmentation method and device and storage medium
CN113516701A (en) Image processing method, image processing device, related equipment and storage medium
CN116993699A (en) Medical image segmentation method and system under eye movement auxiliary training
CN115804579B (en) Process control system for normalized blood pressure measurement and implementation method
CN115082529B (en) System and method for collecting and analyzing multi-dimensional information of gross tissue
CN114783575A (en) Medical image processing system and method
CN112163974A (en) Operation acquisition, learning and sharing method and system
CN112837261B (en) Cell data labeling method and system integrating scanning data and optical image display
JPH10211173A (en) Diagnostic report preparing device
CN116524766A (en) AR medical imaging auxiliary learning system based on innovation teaching aid

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