CN116959712A - Lung adenocarcinoma prognosis method, system, equipment and storage medium based on pathological image - Google Patents
Lung adenocarcinoma prognosis method, system, equipment and storage medium based on pathological image Download PDFInfo
- Publication number
- CN116959712A CN116959712A CN202310941240.0A CN202310941240A CN116959712A CN 116959712 A CN116959712 A CN 116959712A CN 202310941240 A CN202310941240 A CN 202310941240A CN 116959712 A CN116959712 A CN 116959712A
- Authority
- CN
- China
- Prior art keywords
- pathological
- lung adenocarcinoma
- cell
- prognosis
- preset
- 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.)
- Pending
Links
- 230000001575 pathological effect Effects 0.000 title claims abstract description 190
- 238000004393 prognosis Methods 0.000 title claims abstract description 100
- 208000010507 Adenocarcinoma of Lung Diseases 0.000 title claims abstract description 89
- 201000005249 lung adenocarcinoma Diseases 0.000 title claims abstract description 89
- 238000000034 method Methods 0.000 title claims abstract description 48
- 210000004027 cell Anatomy 0.000 claims description 108
- 210000001519 tissue Anatomy 0.000 claims description 58
- 238000012360 testing method Methods 0.000 claims description 32
- 210000003855 cell nucleus Anatomy 0.000 claims description 29
- 238000012549 training Methods 0.000 claims description 20
- 210000000805 cytoplasm Anatomy 0.000 claims description 15
- 238000000605 extraction Methods 0.000 claims description 14
- 210000004072 lung Anatomy 0.000 claims description 12
- 238000003066 decision tree Methods 0.000 claims description 10
- 230000008030 elimination Effects 0.000 claims description 10
- 238000003379 elimination reaction Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 9
- 230000007457 establishment of nucleus localization Effects 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 8
- 230000002708 enhancing effect Effects 0.000 claims description 8
- 230000007170 pathology Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 210000004940 nucleus Anatomy 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 230000001413 cellular effect Effects 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000011426 transformation method Methods 0.000 claims description 3
- 230000030648 nucleus localization Effects 0.000 claims 1
- 230000011218 segmentation Effects 0.000 claims 1
- 238000010801 machine learning Methods 0.000 abstract description 2
- 206010028980 Neoplasm Diseases 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000034994 death Effects 0.000 description 2
- 231100000517 death Toxicity 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 206010058467 Lung neoplasm malignant Diseases 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 231100000504 carcinogenesis Toxicity 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 201000005202 lung cancer Diseases 0.000 description 1
- 208000020816 lung neoplasm Diseases 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- 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
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- 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
- 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
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/54—Extraction of image or video features relating to texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/776—Validation; Performance evaluation
-
- 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
- G06T2207/30061—Lung
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The embodiment of the invention discloses a lung adenocarcinoma prognosis method, a lung adenocarcinoma prognosis system, lung adenocarcinoma prognosis equipment and a lung adenocarcinoma storage medium based on pathological images. Based on the cell characteristics, a machine learning method is used for carrying out accurate prognosis of the lung adenocarcinoma patients, and the prognosis information of the lung adenocarcinoma patients with high accuracy is obtained. The embodiment of the invention solves the technical problems that pathological pictures are difficult to accurately quantify and apply, and has accurate prognosis information and lower cost.
Description
Technical Field
The embodiment of the invention relates to the technical field of image recognition, in particular to a lung adenocarcinoma prognosis method, a lung adenocarcinoma prognosis system, lung adenocarcinoma prognosis equipment and lung adenocarcinoma storage medium based on pathological images.
Background
Lung cancer is the most common cancer worldwide and is also the leading cause of cancer death. About 210 tens of thousands of new cases and 180 tens of deaths occur each year, and lung adenocarcinoma is the predominant histological subtype among them. Histopathological images contain a large amount of information about tumor morphology and its correlation with the surrounding microenvironment, and with the development of accurate oncology, histopathological images have become the gold standard for the diagnostic staging of lung adenocarcinoma.
In the prior art, the analysis of pathological pictures can only be visually perceived through naked eyes, and the visual result cannot be quantified when a large amount of information is ignored, so that the prognosis information of a lung adenocarcinoma patient is difficult to obtain in the current histopathological evaluation mode, such as tumor grading.
Disclosure of Invention
Therefore, the embodiment of the invention provides a lung adenocarcinoma prognosis method, a lung adenocarcinoma prognosis system, lung adenocarcinoma prognosis equipment and lung adenocarcinoma storage medium based on pathological images, so as to solve the technical problem that prognosis information of lung adenocarcinoma patients is difficult to obtain in the current histopathological evaluation mode.
In order to achieve the above object, the embodiment of the present invention provides the following technical solutions:
according to a first aspect of an embodiment of the present invention, there is provided a lung adenocarcinoma prognosis method based on pathological images, characterized in that the method includes:
s1, acquiring lung pathological pictures and prognosis data of a corresponding lung adenocarcinoma patient, dividing the pathological pictures into first pathological sections with preset sizes, and preprocessing the first pathological sections to generate second pathological sections;
s2, dividing the second pathological section by using a preset identification model, positioning a cell tissue position, generating cell tissue positioning mask and cell tissue position information, identifying a cell nucleus from the cell tissue position information by using a preset background elimination model, generating cell nucleus positioning mask and cell nucleus position information, positioning whole cells by combining the cell nucleus position information and the cell tissue position information by using a preset auxiliary identification model, and generating a third pathological section with whole cell position information;
s3, enhancing the image characteristics of the third pathological section by using a black filter and a white filter respectively to generate a fourth pathological section, and carrying out positioning identification on cytoplasm in the fourth pathological section by using a preset auxiliary identification model to generate a cytoplasm positioning mask and cytoplasm position information;
s4, respectively carrying out feature extraction and texture analysis on the cell tissue and the whole cells according to the cell tissue position information and the whole cell position information to generate pathological features of a patient and a pathological feature set of the patient;
s5, constructing a lung adenocarcinoma prognosis model, dividing the pathological feature set of the patient according to a preset proportion, generating a training set and a testing set, training the lung adenocarcinoma prognosis model by using information in the training set, and testing the lung adenocarcinoma prognosis model by using the testing machine;
and S6, inputting a lung pathological image of a patient to be predicted after the current lung adenocarcinoma prognosis model is tested, and generating lung adenocarcinoma prognosis data of the patient when the accuracy rate of the prognosis result of the preset year in the testing machine is higher than the preset information.
Further, collecting prognosis data of a lung pathological image and a corresponding lung adenocarcinoma patient, dividing the pathological image into a first pathological section with a preset size, preprocessing the first pathological section, and generating a second pathological section, wherein the method comprises the following steps:
dividing the pathology into a plurality of first pathological sections with preset sizes;
inputting the first pathological section into a preset pathological texture feature extraction module to generate a black-white first pathological section;
and carrying out brightness correction on the black-white first pathological section by using a preset brightness correction model in sequence to generate a corresponding second pathological section.
Further, the second pathological section is segmented by using a preset recognition model, and a cell tissue position is positioned from the second pathological section, so as to generate a cell tissue positioning mask and cell tissue position information, which comprises the following steps:
performing gradient calculation on the second pathological image, dividing according to gradient calculation data, and positioning the cell tissue position;
performing threshold identification by using an Ojin method, performing threshold processing on the second pathological image, and sequencing gray values of all edge points from low to high, wherein the gray values are cell tissue parts when the gray values are larger than the threshold;
the gradient calculation formula is as follows:
g(x,y)＝grad(f(x,y))＝{[f(x,y)-f(x-1,y)]2[f(x,y)-f(x,y-1)]2}0.5
g(x,y)＝max(grad(f(x,y)),k)
where f (x, y) is an original image, grad represents gradient computation, g (x, y) is gradient computation data, k is a threshold value, and max represents a maximum value.
Further, identifying a nucleus from the tissue location information by using a preset background elimination model, generating a nucleus positioning mask and the nucleus location information, positioning the whole cell by combining the nucleus location information and the tissue location information by using a preset auxiliary identification model, and generating a third pathological section with the whole cell location information, including:
identifying cell nuclei from the cell tissue position information by using a MOG background elimination method, and generating cell nucleus positioning mask and cell nucleus position information;
performing image scanning on the second pathological section to find out a pixel which does not acquire attribution;
taking the pixel as a center, considering a preset neighborhood pixel of the pixel, and combining the neighborhood pixel meeting a growth criterion with the pixel into a pixel region;
traversing all pixels until no pixel without attribution is obtained, wherein the pixel area is a whole cell area;
and generating a third pathological section with whole cell position information by using the whole cell region information.
Further, enhancing the image features of the third pathological section with black and white filters, respectively, to generate a fourth pathological section, comprising:
respectively calculating the open operation and the close operation of the third pathological section to generate a corresponding open operation result and a corresponding close operation result;
generating a top cap operation result by subtracting the open operation result from the third pathological section, and generating a bottom cap operation result by subtracting the third pathological section from the closed operation result;
and enhancing the image characteristics of the third pathological section by using the top cap operation result and the bottom cap operation result, and generating a fourth pathological section.
Further, according to the cell tissue position information and the whole cell position information, respectively performing feature extraction and texture analysis on the cell tissue and the whole cell to generate a patient pathological feature and a patient pathological feature set, including:
extracting a plurality of patient pathological features from the fourth pathological slice by using a gray level symbiotic matrix method, a gray level run-length matrix method, a Wo Luo noe method, filtering and wavelet transformation method respectively;
measuring image brightness according to the gray average value and the gray variance of the fourth pathological section, and measuring adjacent target cells by using the number and the duty ratio of adjacent cells to generate at least one pathological feature of a patient;
extracting the size and shape characteristics of cells from the fourth pathological section to generate patient cell characteristics;
integrating the plurality of patient pathological features with patient cellular features to form a patient pathological feature set.
Further, constructing a lung adenocarcinoma prognosis model, separating the pathological feature sets of the patient according to a preset proportion, generating a training set and a testing set, training the lung adenocarcinoma prognosis model by using information in the training set and testing the lung adenocarcinoma prognosis model by using the testing machine, and the method comprises the following steps:
randomly selecting part of samples and corresponding features, and constructing a preset number of decision trees;
voting weights of all decision trees to generate a prognosis result;
and adjusting weight data of each decision tree until the accuracy of the prognosis result in the test set is higher than a preset threshold.
According to a second aspect of embodiments of the present invention, there is provided a lung adenocarcinoma prognosis system based on pathological images, the system comprising:
the acquisition module is used for acquiring lung pathological pictures and prognosis data of corresponding lung adenocarcinoma patients, dividing the pathological pictures into first pathological sections with preset sizes, preprocessing the first pathological sections, and generating second pathological sections;
the cell positioning module is used for dividing the second pathological section by using a preset identification model, positioning a cell tissue position from the second pathological section to generate cell tissue positioning mask and cell tissue position information, identifying a cell nucleus from the cell tissue position information by using a preset background elimination model, generating cell nucleus positioning mask and cell nucleus position information, positioning whole cells by combining the cell nucleus position information and the cell tissue position information by using a preset auxiliary identification model, and generating a third pathological section with whole cell position information;
the characteristic enhancement module is used for enhancing the image characteristics of the third pathological section by using a black filter and a white filter respectively to generate a fourth pathological section, and carrying out positioning identification on cytoplasm in the fourth pathological section by using a preset auxiliary identification model to generate a cytoplasm positioning mask and cytoplasm position information;
the characteristic extraction module is used for respectively carrying out characteristic extraction and texture analysis on the cell tissues and the whole cells according to the cell tissue position information and the whole cell position information to generate pathological characteristics of a patient and a pathological characteristic set of the patient;
the test module is used for constructing a lung adenocarcinoma prognosis model, separating the pathological feature sets of the patient according to a preset proportion, generating a training set and a test set, training the lung adenocarcinoma prognosis model by using information in the training set and testing the lung adenocarcinoma prognosis model by using the test machine;
and the prognosis module is used for inputting a lung pathological image of a patient to be predicted after the current lung adenocarcinoma prognosis model is tested and generating lung adenocarcinoma prognosis data of the patient when the accuracy rate of the prognosis result of the preset year in the testing machine is higher than the preset information.
According to a third aspect of embodiments of the present invention, there is provided a lung adenocarcinoma prognosis apparatus based on pathological images, the apparatus comprising: a processor and a memory;
the memory is used for storing one or more program instructions;
the processor is configured to execute one or more program instructions to perform the steps of the pathology image-based lung adenocarcinoma prognosis method according to any of the preceding claims.
According to a fourth aspect of embodiments of the present invention, there is provided a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the pathological image based lung adenocarcinoma prognosis method as described in any of the above.
The embodiment of the invention has the following advantages:
after receiving a pathological picture of a patient, the embodiment of the invention analyzes the basic characteristics and the higher-order characteristics of cells of lung adenocarcinoma by an automatic method. Based on the cell characteristics, a machine learning method is used for carrying out accurate prognosis of the lung adenocarcinoma patients, and the prognosis information of the lung adenocarcinoma patients with high accuracy is obtained. The embodiment of the invention solves the technical problems that pathological pictures are difficult to accurately quantify and apply, and has accurate prognosis information and lower cost.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It will be apparent to those skilled in the art from this disclosure that the drawings described below are merely exemplary and that other embodiments may be derived from the drawings provided without undue effort.
The structures, proportions, sizes, etc. shown in the present specification are shown only for the purposes of illustration and description, and are not intended to limit the scope of the invention, which is defined by the claims, so that any structural modifications, changes in proportions, or adjustments of sizes, which do not affect the efficacy or the achievement of the present invention, should fall within the ambit of the technical disclosure.
FIG. 1 is a schematic diagram of a lung adenocarcinoma prognosis system based on pathological images according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a lung adenocarcinoma prognosis method based on pathological images according to an embodiment of the invention.
Detailed Description
Other advantages and advantages of the present invention will become apparent to those skilled in the art from the following detailed description, which, by way of illustration, is to be read in connection with certain specific embodiments, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the prior art, the analysis of pathological pictures can only be visually perceived through naked eyes, and the visual result cannot be quantified when a large amount of information is ignored, so that the prognosis information of a lung adenocarcinoma patient is difficult to obtain in the current histopathological evaluation mode, such as tumor grading.
In order to solve the technical problem that the prognosis information of the lung adenocarcinoma patients is difficult to obtain in the histopathological evaluation mode.
Referring to fig. 1, an embodiment of the present invention discloses a lung adenocarcinoma prognosis system based on pathological images, comprising: an acquisition module 1; a cell positioning module 2; a feature enhancement module 3; a feature extraction module 4; a test module 5; prognosis module 6.
Corresponding to the lung adenocarcinoma prognosis system based on the pathological image disclosed above, the embodiment of the invention also discloses a lung adenocarcinoma prognosis method based on the pathological image. The lung adenocarcinoma prognosis method based on the pathological image disclosed in the embodiment of the invention is described in detail below in conjunction with the lung adenocarcinoma prognosis system based on the pathological image described above.
Referring to fig. 2, an acquisition module 1 acquires a lung pathological image and prognosis data of a corresponding lung adenocarcinoma patient, divides the pathological image into first pathological sections of a preset size, and performs pretreatment on the first pathological sections to generate second pathological sections.
Dividing the pathology into a plurality of first pathological sections with preset sizes; inputting the first pathological section into a preset pathological texture feature extraction module to generate a black-white first pathological section; and carrying out brightness correction on the black-white first pathological section by using a preset brightness correction model in sequence to generate a corresponding second pathological section.
The preset size is 1000×1000 pixels of pathological sections, and the brightness correction is to normalize the brightness of each section to a range of 0-1.
The cell positioning module 2 uses a preset recognition model to divide the second pathological section, positions the cell tissue from the second pathological section to generate cell tissue positioning mask and cell tissue position information, uses a preset background elimination model to recognize the cell nucleus from the cell tissue position information to generate cell nucleus positioning mask and cell nucleus position information, and uses a preset auxiliary recognition model to combine the cell nucleus position information and the cell tissue position information to position the whole cell so as to generate a third pathological section with the whole cell position information.
Performing gradient calculation on the second pathological image, dividing according to gradient calculation data, and positioning the cell tissue position; and carrying out threshold identification by using an Ojin method, carrying out threshold processing on the second pathological image, and sequencing gray values of all edge points from low to high, wherein the gray values are the cell tissue parts when the gray values are larger than the threshold.
The gradient calculation formula is as follows:
g(x,y)＝grad(f(x,y))＝{[f(x,y)-f(x-1,y)]2[f(x,y)-f(x,y-1)]2}0.5
g(x,y)＝max(grad(f(x,y)),k)
where f (x, y) is an original image, grad represents gradient computation, g (x, y) is gradient computation data, k is a threshold value, and max represents a maximum value.
Wherein, the pixel is selected to be 50-800, the identification threshold method is selected to be the Ojin method, and the edge threshold is set to be 15.
Identifying cell nuclei from the cell tissue position information by using a MOG background elimination method, and generating cell nucleus positioning mask and cell nucleus position information; performing image scanning on the second pathological section to find out a pixel which does not acquire attribution; taking the pixel as a center, considering a preset neighborhood pixel of the pixel, and combining the neighborhood pixel meeting a growth criterion with the pixel into a pixel region; traversing all pixels until no pixel without attribution is obtained, wherein the pixel area is a whole cell area; and generating a third pathological section with whole cell position information by using the whole cell region information.
Wherein, the pixel is selected to be 10-30, the correction factor is adjusted to be 0.8-1.2, and the overlapping parameter is excluded to be the intensity.
In the whole cell identification, the identification mode is a propagation mode, the threshold method is selected as the Ojin method, and the smoothing parameter is set to 0.4.
The characteristic enhancement module 3 enhances the image characteristics of the third pathological section by using a black filter and a white filter respectively to generate a fourth pathological section, and performs positioning identification on cytoplasm in the fourth pathological section by using a preset auxiliary identification model to generate a cytoplasm positioning mask and cytoplasm position information.
Respectively calculating the open operation and the close operation of the third pathological section to generate a corresponding open operation result and a corresponding close operation result; generating a top cap operation result by subtracting the open operation result from the third pathological section, and generating a bottom cap operation result by subtracting the third pathological section from the closed operation result; and enhancing the image characteristics of the third pathological section by using the top cap operation result and the bottom cap operation result, and generating a fourth pathological section.
And the feature extraction module 4 respectively performs feature extraction and texture analysis on the cell tissues and the whole cells according to the cell tissue position information and the whole cell position information to generate pathological features of a patient and a pathological feature set of the patient.
Extracting a plurality of patient pathological features from the fourth pathological slice by using a gray level symbiotic matrix method, a gray level run-length matrix method, a Wo Luo noe method, filtering and wavelet transformation method respectively; measuring image brightness according to the gray average value and the gray variance of the fourth pathological section, and measuring adjacent target cells by using the number and the duty ratio of adjacent cells to generate at least one pathological feature of a patient; extracting the size and shape characteristics of cells from the fourth pathological section to generate patient cell characteristics; integrating the plurality of patient pathological features with patient cellular features to form a patient pathological feature set.
And the test module 5 constructs a lung adenocarcinoma prognosis model, separates the pathological feature sets of the patient according to a preset proportion, generates a training set and a test set, trains the lung adenocarcinoma prognosis model by using information in the training set and tests the lung adenocarcinoma prognosis model by using the test machine.
Randomly selecting part of samples and corresponding features, and constructing a preset number of decision trees; voting weights of all decision trees to generate a prognosis result; and adjusting weight data of each decision tree until the accuracy of the prognosis result in the test set is higher than a preset threshold.
Wherein the preset ratio is 8:2, setting the size of the nodes as 12, setting the number of decision trees as 1000, and setting the parameters as permutation.
And the prognosis module 6 inputs a lung pathological image of a patient to be predicted to generate lung adenocarcinoma prognosis data of the patient after the current lung adenocarcinoma prognosis model is tested when the accuracy rate of the prognosis result of the preset year in the testing machine is higher than the preset information.
Wherein the preset years are the first year, the third year and the fifth year of the patient.
In addition, the embodiment of the invention also provides lung adenocarcinoma prognosis equipment based on pathological images, which comprises the following steps: a processor and a memory; the memory is used for storing one or more program instructions; the processor is configured to execute one or more program instructions to perform the steps of the pathology image-based lung adenocarcinoma prognosis method according to any of the preceding claims.
In addition, the embodiment of the invention also provides a computer readable storage medium, wherein a computer program is stored on the computer readable storage medium, and the computer program realizes the steps of the lung adenocarcinoma prognosis method based on the pathological image when being executed by a processor.
In the embodiment of the invention, the processor may be an integrated circuit chip with signal processing capability. The processor may be a general purpose processor, a digital signal processor (Digital Signal Processor, DSP for short), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC for short), a field programmable gate array (FieldProgrammable GateArray, FPGA for short), or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.
The disclosed methods, steps, and logic blocks in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present invention may be embodied directly in the execution of a hardware decoding processor, or in the execution of a combination of hardware and software modules in a decoding processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. The processor reads the information in the storage medium and, in combination with its hardware, performs the steps of the above method.
The storage medium may be memory, for example, may be volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory.
The nonvolatile Memory may be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an electrically Erasable ROM (Electrically EPROM, EEPROM), or a flash Memory.
The volatile memory may be a random access memory (Random Access Memory, RAM for short) which acts as an external cache. By way of example, and not limitation, many forms of RAM are available, such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (Double Data RateSDRAM), enhanced SDRAM (ESDRAM), synchronous DRAM (SLDRAM), and direct memory bus RAM (directracram, DRRAM).
The storage media described in embodiments of the present invention are intended to comprise, without being limited to, these and any other suitable types of memory.
Those skilled in the art will appreciate that in one or more of the examples described above, the functions described in the present invention may be implemented in a combination of hardware and software. When the software is applied, the corresponding functions may be stored in a computer-readable medium or transmitted as one or more instructions or code on the computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media may be any available media that can be accessed by a general purpose or special purpose computer.
While the invention has been described in detail in the foregoing general description and specific examples, it will be apparent to those skilled in the art that modifications and improvements can be made thereto. Accordingly, such modifications or improvements may be made without departing from the spirit of the invention and are intended to be within the scope of the invention as claimed.
Claims (10)
1. A method of prognosis of lung adenocarcinoma based on pathological images, characterized in that it comprises:
s1, acquiring lung pathological pictures and prognosis data of a corresponding lung adenocarcinoma patient, dividing the pathological pictures into first pathological sections with preset sizes, and preprocessing the first pathological sections to generate second pathological sections;
s2, dividing the second pathological section by using a preset identification model, positioning a cell tissue position, generating cell tissue positioning mask and cell tissue position information, identifying a cell nucleus from the cell tissue position information by using a preset background elimination model, generating cell nucleus positioning mask and cell nucleus position information, positioning whole cells by combining the cell nucleus position information and the cell tissue position information by using a preset auxiliary identification model, and generating a third pathological section with whole cell position information;
s3, enhancing the image characteristics of the third pathological section by using a black filter and a white filter respectively to generate a fourth pathological section, and carrying out positioning identification on cytoplasm in the fourth pathological section by using a preset auxiliary identification model to generate a cytoplasm positioning mask and cytoplasm position information;
s4, respectively carrying out feature extraction and texture analysis on the cell tissue and the whole cells according to the cell tissue position information and the whole cell position information to generate pathological features of a patient and a pathological feature set of the patient;
s5, constructing a lung adenocarcinoma prognosis model, dividing the pathological feature set of the patient according to a preset proportion, generating a training set and a testing set, training the lung adenocarcinoma prognosis model by using information in the training set, and testing the lung adenocarcinoma prognosis model by using the testing machine;
and S6, inputting a lung pathological image of a patient to be predicted after the current lung adenocarcinoma prognosis model is tested, and generating lung adenocarcinoma prognosis data of the patient when the accuracy rate of the prognosis result of the preset year in the testing machine is higher than the preset information.
2. The pathological-image-based lung adenocarcinoma prognosis method of claim 1, wherein acquiring lung pathological pictures and prognosis data of a corresponding lung adenocarcinoma patient, dividing the pathological pictures into first pathological sections of a preset size and preprocessing the first pathological sections to generate second pathological sections, comprises:
dividing the pathology into a plurality of first pathological sections with preset sizes;
inputting the first pathological section into a preset pathological texture feature extraction module to generate a black-white first pathological section;
and carrying out brightness correction on the black-white first pathological section by using a preset brightness correction model in sequence to generate a corresponding second pathological section.
3. The pathological-image-based lung adenocarcinoma prognosis method according to claim 2, wherein the segmentation of the second pathological section using a preset recognition model from which a cell tissue position is located, generating a cell tissue positioning mask and cell tissue position information, comprises:
performing gradient calculation on the second pathological image, dividing according to gradient calculation data, and positioning the cell tissue position;
performing threshold identification by using an Ojin method, performing threshold processing on the second pathological image, and sequencing gray values of all edge points from low to high, wherein the gray values are cell tissue parts when the gray values are larger than the threshold;
the gradient calculation formula is as follows:
g(x,y)＝grad(f(x,y))＝{[f(x,y)-f(x-1,y)]2[f(x,y)-f(x,y-1)]2}0.5
g(x,y)＝max(grad(f(x,y)),k)
where f (x, y) is an original image, grad represents gradient computation, g (x, y) is gradient computation data, k is a threshold value, and max represents a maximum value.
4. The pathological-image-based lung adenocarcinoma prognosis method of claim 3, wherein the steps of identifying nuclei from the tissue location information using a preset background elimination model, generating a nuclear localization mask and the nuclear location information, locating whole cells using a preset auxiliary identification model in combination with the nuclear location information and the tissue location information, and generating a third pathological section having the whole-cell location information include:
identifying cell nuclei from the cell tissue position information by using a MOG background elimination method, and generating cell nucleus positioning mask and cell nucleus position information;
performing image scanning on the second pathological section to find out a pixel which does not acquire attribution;
taking the pixel as a center, considering a preset neighborhood pixel of the pixel, and combining the neighborhood pixel meeting a growth criterion with the pixel into a pixel region;
traversing all pixels until no pixel without attribution is obtained, wherein the pixel area is a whole cell area;
and generating a third pathological section with whole cell position information by using the whole cell region information.
5. The pathological-image-based lung adenocarcinoma prognosis method of claim 4, wherein the image features of the third pathological section are enhanced by using black and white filters, respectively, and generating a fourth pathological section includes:
respectively calculating the open operation and the close operation of the third pathological section to generate a corresponding open operation result and a corresponding close operation result;
generating a top cap operation result by subtracting the open operation result from the third pathological section, and generating a bottom cap operation result by subtracting the third pathological section from the closed operation result;
and enhancing the image characteristics of the third pathological section by using the top cap operation result and the bottom cap operation result, and generating a fourth pathological section.
6. The pathological-image-based lung adenocarcinoma prognosis method according to claim 5, wherein the feature extraction and texture analysis are performed on the cell tissue and the whole cell according to the cell tissue position information and the whole-cell position information, respectively, to generate a patient pathological feature and a patient pathological feature set, comprising:
extracting a plurality of patient pathological features from the fourth pathological slice by using a gray level symbiotic matrix method, a gray level run-length matrix method, a Wo Luo noe method, filtering and wavelet transformation method respectively;
measuring image brightness according to the gray average value and the gray variance of the fourth pathological section, and measuring adjacent target cells by using the number and the duty ratio of adjacent cells to generate at least one pathological feature of a patient;
extracting the size and shape characteristics of cells from the fourth pathological section to generate patient cell characteristics;
integrating the plurality of patient pathological features with patient cellular features to form a patient pathological feature set.
7. The pathology image-based lung adenocarcinoma prognosis method of claim 6, wherein constructing a lung adenocarcinoma prognosis model, separating the patient pathology feature sets according to a preset ratio, generating a training set and a test set, training the lung adenocarcinoma prognosis model using information in the training set and testing the lung adenocarcinoma prognosis model using the tester, comprises:
randomly selecting part of samples and corresponding features, and constructing a preset number of decision trees;
voting weights of all decision trees to generate a prognosis result;
and adjusting weight data of each decision tree until the accuracy of the prognosis result in the test set is higher than a preset threshold.
8. A lung adenocarcinoma prognosis system based on pathological images, characterized in that it comprises:
the acquisition module is used for acquiring lung pathological pictures and prognosis data of corresponding lung adenocarcinoma patients, dividing the pathological pictures into first pathological sections with preset sizes, preprocessing the first pathological sections, and generating second pathological sections;
the cell positioning module is used for dividing the second pathological section by using a preset identification model, positioning a cell tissue position from the second pathological section to generate cell tissue positioning mask and cell tissue position information, identifying a cell nucleus from the cell tissue position information by using a preset background elimination model, generating cell nucleus positioning mask and cell nucleus position information, positioning whole cells by combining the cell nucleus position information and the cell tissue position information by using a preset auxiliary identification model, and generating a third pathological section with whole cell position information;
the characteristic enhancement module is used for enhancing the image characteristics of the third pathological section by using a black filter and a white filter respectively to generate a fourth pathological section, and carrying out positioning identification on cytoplasm in the fourth pathological section by using a preset auxiliary identification model to generate a cytoplasm positioning mask and cytoplasm position information;
the characteristic extraction module is used for respectively carrying out characteristic extraction and texture analysis on the cell tissues and the whole cells according to the cell tissue position information and the whole cell position information to generate pathological characteristics of a patient and a pathological characteristic set of the patient;
the test module is used for constructing a lung adenocarcinoma prognosis model, separating the pathological feature sets of the patient according to a preset proportion, generating a training set and a test set, training the lung adenocarcinoma prognosis model by using information in the training set and testing the lung adenocarcinoma prognosis model by using the test machine;
and the prognosis module is used for inputting a lung pathological image of a patient to be predicted after the current lung adenocarcinoma prognosis model is tested and generating lung adenocarcinoma prognosis data of the patient when the accuracy rate of the prognosis result of the preset year in the testing machine is higher than the preset information.
9. A lung adenocarcinoma prognosis apparatus based on pathological images, characterized in that it comprises: a processor and a memory;
the memory is used for storing one or more program instructions;
the processor for executing one or more program instructions for performing the steps of the pathology image-based lung adenocarcinoma prognosis method according to any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, which computer program, when being executed by a processor, implements the steps of the pathological-image-based lung adenocarcinoma prognosis method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310941240.0A CN116959712A (en) | 2023-07-28 | 2023-07-28 | Lung adenocarcinoma prognosis method, system, equipment and storage medium based on pathological image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310941240.0A CN116959712A (en) | 2023-07-28 | 2023-07-28 | Lung adenocarcinoma prognosis method, system, equipment and storage medium based on pathological image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116959712A true CN116959712A (en) | 2023-10-27 |
Family
ID=88448844
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310941240.0A Pending CN116959712A (en) | 2023-07-28 | 2023-07-28 | Lung adenocarcinoma prognosis method, system, equipment and storage medium based on pathological image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116959712A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117423479A (en) * | 2023-12-19 | 2024-01-19 | 神州医疗科技股份有限公司 | Prediction method and system based on pathological image data |
-
2023
- 2023-07-28 CN CN202310941240.0A patent/CN116959712A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117423479A (en) * | 2023-12-19 | 2024-01-19 | 神州医疗科技股份有限公司 | Prediction method and system based on pathological image data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110110799B (en) | Cell sorting method, cell sorting device, computer equipment and storage medium | |
| CN111723860B (en) | Target detection method and device | |
| CN109886179B (en) | Image segmentation method and system of cervical cell smear based on Mask-RCNN | |
| CN108665456B (en) | Method and system for real-time marking of breast ultrasound lesion region based on artificial intelligence | |
| CN110111344B (en) | Pathological section image grading method and device, computer equipment and storage medium | |
| CN109886282A (en) | Method for checking object, device, computer readable storage medium and computer equipment | |
| CN110969204B (en) | Sample classification system based on fusion of magnetic resonance image and digital pathology image | |
| CN106056118A (en) | Recognition and counting method for cells | |
| CN112634261A (en) | Stomach cancer focus detection method and device based on convolutional neural network | |
| CN109978037A (en) | Image processing method, model training method, device and storage medium | |
| CN111028923B (en) | Digital pathological image staining normalization method, electronic device and storage medium | |
| CN111931751B (en) | Deep learning training method, target object identification method, system and storage medium | |
| CN112750121B (en) | System and method for detecting digital image quality of pathological slide | |
| CN111161272B (en) | Embryo tissue segmentation method based on generation of confrontation network | |
| CN109919254B (en) | Breast density classification method, system, readable storage medium and computer device | |
| CN111784721A (en) | Ultrasonic endoscopic image intelligent segmentation and quantification method and system based on deep learning | |
| CN112651979A (en) | Lung X-ray image segmentation method, system, computer equipment and storage medium | |
| CN110853009A (en) | Retina pathology image analysis system based on machine learning | |
| CN116959712A (en) | Lung adenocarcinoma prognosis method, system, equipment and storage medium based on pathological image | |
| CN112215217B (en) | Digital image recognition method and device for simulating doctor to read film | |
| CN115760858B (en) | Deep learning-based kidney pathological section cell identification method and system | |
| CN114841947A (en) | Method and device for multi-scale feature extraction and prognosis analysis of H & E staining pathological image tumor region | |
| CN113160185A (en) | Method for guiding cervical cell segmentation by using generated boundary position | |
| CN114742758A (en) | Cell nucleus classification method in full-field digital slice histopathology picture | |
| CN114549462A (en) | Focus detection method, device, equipment and medium based on visual angle decoupling Transformer model |
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 |