CN111105407A

CN111105407A - Pathological section staining quality evaluation method, device, equipment and storage medium

Info

Publication number: CN111105407A
Application number: CN201911354095.6A
Authority: CN
Inventors: 车拴龙; 罗丕福; 钟学军; 徐炜; 苏丽珠; 林万里; 陈蝶; 丁向东
Original assignee: Guangzhou Kingmed Diagnostics Central Co Ltd
Current assignee: Guangzhou Kingmed Diagnostics Central Co Ltd
Priority date: 2019-12-25
Filing date: 2019-12-25
Publication date: 2020-05-05
Anticipated expiration: 2039-12-25
Also published as: CN111105407B

Abstract

The embodiment of the invention discloses a pathological section staining quality evaluation method, a pathological section staining quality evaluation device, pathological section staining quality evaluation equipment and a storage medium. The method comprises the following steps: carrying out digital scanning on the dyed pathological section to obtain a medical image to be analyzed; performing feature recognition on the medical image to be analyzed to obtain a plurality of dyeing color values; and acquiring a dyeing standard value, and determining the dyeing quality of the pathological section corresponding to the medical image to be analyzed according to the plurality of dyeing color values and the dyeing standard value. The evaluation standard of the invention is uniform, the subjective evaluation of pathologists or pathological technicians is avoided by adopting artificial intelligence, the consistency of quality evaluation is ensured, and the invention is suitable for the pathological section staining quality evaluation in the same laboratory and the pathological section staining quality evaluation between different laboratories.

Description

Pathological section staining quality evaluation method, device, equipment and storage medium

Technical Field

The invention relates to the technical field of quality evaluation, in particular to a method, a device, equipment and a storage medium for evaluating the staining quality of a pathological section.

Background

The pathological technique is an important branch of pathology, is a methodology in pathological research and is the basis of pathological diagnosis. Conventional pathology is the most important part of the pathological technique, and any pathological diagnosis cannot be separated from it. The immunohistochemical staining is an indispensable important technology for modern pathological diagnosis technology. The steps of immunohistochemical staining are complicated, and the final immunohistochemical staining effect may be completely incomplete due to the operation techniques of different laboratories and technicians. In the same laboratory, three different effects of staining appear after 3 laboratory operators operate at different times, and the result of the later pathological diagnosis report can be seriously affected.

At present, the quality control in the same laboratory or among different laboratories is usually subjectively evaluated by a pathologist or a pathologist. On the one hand, the criteria for diagnostic quality are not completely consistent for everyone, and most are subjective and difficult to repeat evaluations; on the other hand, for slight quality differences, the human evaluation results are often more different, and reliable and accurate evaluation cannot be completed. Therefore, the quality control evaluation standard and result of each time are very unstable, so that the quality of immunohistochemical staining is difficult to achieve consistent staining quality in different laboratories. Therefore, it is important to develop a method for evaluating the staining quality of pathological sections, which has uniform detection standards and is suitable for the same laboratory and different laboratories.

Disclosure of Invention

In view of the above, it is necessary to provide a pathological section staining quality evaluation method, apparatus, device, and storage medium.

In a first aspect, the invention provides a method for evaluating staining quality of a pathological section, comprising the following steps:

carrying out digital scanning on the dyed pathological section to obtain a medical image to be analyzed;

performing feature recognition on the medical image to be analyzed to obtain a plurality of dyeing color values;

and acquiring a dyeing standard value, and determining the dyeing quality of the pathological section corresponding to the medical image to be analyzed according to the plurality of dyeing color values and the dyeing standard value.

In a second aspect, the present invention further provides a device for evaluating the staining quality of a pathological section, the device comprising:

the scanning module is used for digitally scanning the dyed pathological section to obtain a medical image to be analyzed;

the dyeing color value extraction module is used for performing feature recognition on the medical image to be analyzed to obtain a plurality of dyeing color values;

and the dyeing quality determining module is used for acquiring a dyeing standard value and determining the dyeing quality of the pathological section corresponding to the medical image to be analyzed according to the plurality of dyeing color values and the dyeing standard value.

In a third aspect, the present invention also provides a storage medium storing a computer program of instructions, which, when executed by a processor, causes the processor to perform the steps of the method according to any one of the first aspect.

In a fourth aspect, the present invention also proposes a computer device comprising at least one memory storing a computer program of instructions, at least one processor, which, when executed by the processor, causes the processor to carry out the steps of the method of any one of the first aspects.

In conclusion, the pathological section staining quality evaluation method digitally scans the stained pathological section to obtain the medical image to be analyzed, and analyzes the digitized medical image to be analyzed through artificial intelligence, so that the subjective evaluation of a pathologist or a pathological technician is avoided, and the consistency of quality evaluation is ensured; treat the medical image of analysis and carry out feature recognition, obtain a plurality of dyeing colour values, acquire the dyeing standard value, confirm the dyeing quality with the pathological section that treats analysis medical image and correspond according to a plurality of dyeing colour values and dyeing standard value, the method of discerning the dyeing colour value is the same, adopt a plurality of dyeing colour values and dyeing standard value to evaluate, the evaluation parameter is unified, the evaluation standard is unified, adopt artificial intelligence, thereby be fit for pathological section dyeing quality evaluation in the same laboratory, also be fit for pathological section dyeing quality evaluation between different laboratories. Therefore, the evaluation standard of the invention is uniform, the subjective evaluation of pathologists or pathological technicians is avoided by adopting artificial intelligence, the consistency of quality evaluation is ensured, and the invention is suitable for the pathological section staining quality evaluation in the same laboratory and the pathological section staining quality evaluation between different laboratories.

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 is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Wherein:

FIG. 1 is a flowchart of a method for evaluating staining quality of a pathological section according to an embodiment;

FIG. 2 is a flow chart of the method for evaluating staining quality of a pathological section of FIG. 1 for identifying staining color values;

FIG. 3 is a flow chart of the cell structure feature identification of the method for assessing staining quality of a pathological section of FIG. 1;

FIG. 4 is a flow chart of the method for evaluating staining quality of a pathological section of FIG. 1;

FIG. 5 is a flow chart of the method for evaluating staining quality of a pathological section of FIG. 1 for determining staining quality;

FIG. 6 is a flowchart of the method for evaluating the staining quality of a pathological section of FIG. 1 for calculating a difference in staining color;

FIG. 7 is a flowchart of a method for evaluating staining quality of a pathological section according to another embodiment;

FIG. 8 is a flowchart of the method for evaluating the staining quality of a pathological section of FIG. 1 for analyzing the cause of disqualification;

FIG. 9 is a block diagram showing the structure of a staining quality evaluating apparatus for pathological section in one embodiment;

FIG. 10 is a block diagram showing a configuration of a computer device according to an embodiment.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

As shown in fig. 1, in one embodiment, a method for evaluating staining quality of a pathological section is provided, the method comprising:

s102, carrying out digital scanning on the dyed pathological section to obtain a medical image to be analyzed;

the pathological section is prepared by taking pathological tissues with a certain size, preparing pathological sections by a histopathology method, and further examining pathological changes by a microscope.

Traditional staining of pathological sections includes: HE staining, bus staining, giemsa reynold staining, collagen fiber staining (Masson et al), reticular fiber staining, spandex staining, muscle tissue staining (hematoxylin phosphotungstate), fat staining (sudan III), glycogen staining (PAS), mucus staining (PAS), etc., and immunochemical staining of pathological sections includes: immunohistochemical staining (immunohistochemistry), immunofluorescent staining, and the like, and examples thereof are not particularly limited. The present application is exemplified as applied to immunohistochemistry, it being understood that the methods of the present application are equally applicable to traditional staining of pathological sections and other immunochemical staining.

The digital scanning refers to digital scanning of the stained pathological section to obtain a digital initial medical image. The digital scanning may be implemented by a digital camera or a scanner, which is not limited in this example.

The initial medical image may be directly used as a medical image to be analyzed, or an image obtained by performing modification processing on the initial medical image may be used as the medical image to be analyzed.

S104, performing feature recognition on the medical image to be analyzed to obtain a plurality of dyeing color values;

performing cell structure feature recognition according to the medical image to be analyzed to obtain a cell nucleus image and a cell cytoplasm image; and respectively calculating color values according to the cell nucleus image and the cell pulp image to obtain a plurality of dyeing color values.

It is understood that a plurality of the staining color values may simultaneously include a cell nucleus negative staining color value, a cell cytoplasm negative staining color value, a cell nucleus positive staining color value, a cell nucleus negative staining color value, a cell cytoplasm positive staining color value, a cell nucleus negative staining color value, a cell nucleus positive staining color value, and a cell cytoplasm positive staining color value.

Optionally, the color value refers to a numerical value of three red (R), green (G), and blue (B) color channels, also called RGB value, which is expressed as (Rx, Gx, Bx), where Rx is a numerical value of a red color channel, Gx is a numerical value of a green color channel, and Bx is a numerical value of a blue color channel, which is not limited in this example.

The cell nucleus negative staining color value refers to a color value obtained by negative staining of the cell nucleus.

The cytoplasm negative staining color value refers to a color value obtained by negative staining of cytoplasm.

The cell nucleus positive staining color value refers to a color value obtained by staining the cell nucleus positively.

The cytoplasm positive staining color value refers to a color value obtained by staining cytoplasm positively.

The cell nucleus refers to the largest and most important cell structure in the eukaryotic cell, is a control center of cytogenetics and metabolism, and is one of the most obvious marks for distinguishing the eukaryotic cell from the prokaryotic cell.

The cytoplasm refers to cytoplasm and is a general term for all translucent, colloidal and granular substances surrounded by cytoplasmic membranes except a nuclear region.

S106, obtaining a dyeing standard value, and determining the dyeing quality of the pathological section corresponding to the medical image to be analyzed according to the plurality of dyeing color values and the dyeing standard value.

Alternatively, the standard dyeing value may be a specific RGB value, or may be a range value.

Optionally, the staining color value and a staining standard value corresponding to the staining color value are scored to obtain a staining color value score, comprehensive scoring is performed according to all the staining color value scores to obtain a pathological section staining comprehensive score, a staining comprehensive score standard value is obtained, and the staining quality of a pathological section corresponding to the medical image to be analyzed is determined by comparing the staining comprehensive score standard value and the pathological section staining comprehensive score.

Optionally, when the dyeing color value is within the range of the dyeing standard value corresponding to the dyeing color value, the dyeing color value evaluation result is qualified, and when the dyeing color value is outside the range of the dyeing standard value corresponding to the dyeing color value, the dyeing color value evaluation result is unqualified; and comprehensively evaluating according to all the dyeing color value evaluation results to obtain the dyeing quality of the pathological section corresponding to the medical image to be analyzed.

And comprehensively evaluating according to all the dyeing color value evaluation results to obtain the dyeing quality of the pathological section corresponding to the medical image to be analyzed, wherein the dyeing quality comprises the following steps: when all the dyeing color value evaluation results are qualified, the dyeing quality of the pathological section corresponding to the medical image to be analyzed is qualified, and when all the dyeing color value evaluation results are partially unqualified, the dyeing quality of the pathological section corresponding to the medical image to be analyzed is unqualified; it is understood that other comprehensive evaluation methods can be used, and the examples are not limited to the specific examples.

The standard dyeing value is a color value obtained by digital scanning of a pathological section which is evaluated by an expert pathologist and has the best dyeing quality, and is a reference standard which is obtained by repeated tests and can be used for evaluating the dyeing quality.

It is understood that the number of the dyeing standard values is multiple, and the dyeing standard values correspond to the dyeing color values one by one. For example, the cell nucleus negative staining color value corresponds to a staining standard value, the cell cytoplasm negative staining color value corresponds to a staining standard value, the cell nucleus positive staining color value corresponds to a staining standard value, and the cell cytoplasm positive staining color value corresponds to a staining standard value, which is not specifically limited herein.

Optionally, the staining standard values include: a standard value for cell nucleus negative staining corresponding to the color value for cell nucleus negative staining, a standard value for cell cytoplasm negative staining corresponding to the color value for cell cytoplasm negative staining, a standard value for cell nucleus positive staining corresponding to the color value for cell nucleus positive staining, and a standard value for cell cytoplasm positive staining corresponding to the color value for cell cytoplasm positive staining.

Optionally, the one-to-one correspondence may refer to that the staining standard value and the staining color value corresponding to the staining standard value adopt the same staining method, or that the staining standard value and the staining color value corresponding to the staining standard value adopt the same organ tissue and the same staining method, or that the staining standard value and the staining color value corresponding to the staining standard value adopt the same organ tissue, the same staining method and the same pathology, which is not specifically limited in this example.

Optionally, the stained pathological section refers to a pathological section that completes all staining stages, and the staining standard value is also a staining standard value corresponding to the pathological section that completes all staining stages.

Optionally, the staining includes a plurality of staining stages, and if the pathological section is stained in one of the staining stages, the staining standard value is also the staining standard value corresponding to the pathological section stained in the staining stage. By adopting the staged dyeing quality evaluation of the dyed pathological section, the dyeing stage of the unqualified dyed pathological section can be quickly positioned, and the corresponding problems can be conveniently solved in time in the follow-up process.

According to the pathological section staining quality evaluation method, the stained pathological sections are digitally scanned to obtain medical images to be analyzed, and the digitized medical images to be analyzed are analyzed through artificial intelligence, so that subjective evaluation of pathologists or pathological technicians is avoided, and consistency of quality evaluation is ensured; treat the medical image of analysis and carry out feature recognition, obtain a plurality of dyeing colour values, acquire the dyeing standard value, confirm the dyeing quality with the pathological section that treats analysis medical image and correspond according to a plurality of dyeing colour values and dyeing standard value, the method of discerning the dyeing colour value is the same, adopt a plurality of dyeing colour values and dyeing standard value to evaluate, the evaluation parameter is unified, the evaluation standard is unified, adopt artificial intelligence, thereby be fit for pathological section dyeing quality evaluation in the same laboratory, also be fit for pathological section dyeing quality evaluation between different laboratories.

As shown in fig. 2, in an embodiment, the performing feature recognition on the medical image to be analyzed to obtain a plurality of staining color values includes:

s202, identifying cell structure characteristics according to the medical image to be analyzed to obtain a cell nucleus image and a cell cytoplasm image;

in the prior art, a cell nucleus segmentation algorithm and a cell cytoplasm segmentation algorithm can be selected to respectively identify the characteristics of the cell structure in the medical image to be analyzed, so that a cell nucleus image and a cell cytoplasm image are obtained. It is understood that other algorithms can be used to perform feature recognition on the cell structure in the medical image to be analyzed, which is not specifically limited by the examples.

S204, color values are calculated according to the cell nucleus image and the cell pulp image respectively to obtain a plurality of staining color values, and the staining color values comprise cell nucleus negative staining color values, cell pulp negative staining color values, cell nucleus positive staining color values and/or cell pulp positive staining color values.

As shown in fig. 3, in an embodiment, the performing cell structural feature recognition according to the medical image to be analyzed to obtain a cell nucleus image and a cell cytoplasm image includes:

s302, segmenting cell nucleuses according to the medical image to be analyzed by adopting a cell nucleus segmentation algorithm to obtain cell nucleus images;

specifically, cell nucleuses are segmented by adopting a cell nucleus segmentation algorithm according to the medical image to be analyzed to obtain cell nucleus images, and the number of the cell nucleus images is multiple. The cell nucleus segmentation algorithm may be selected from the prior art and will not be described herein.

S304, segmenting cytoplasm by adopting a cytoplasm segmentation algorithm according to the medical image to be analyzed to obtain a cytoplasm image.

Specifically, cytoplasm is segmented by adopting a cytoplasm segmentation algorithm according to the medical image to be analyzed to obtain cytoplasm images, and the number of the cytoplasm images is multiple. The cytoplasm segmentation algorithm can be selected from the prior art and is not described in detail herein.

As shown in fig. 4, in an embodiment, the performing color value calculation according to the cell nucleus image and the cell cytoplasm image respectively to obtain a plurality of staining color values, where the plurality of staining color values include a cell nucleus negative staining color value, a cell cytoplasm negative staining color value, a cell nucleus positive staining color value and/or a cell cytoplasm positive staining color value includes:

s402, classifying according to the color values of the cell nucleus images to obtain a cell nucleus negative staining image and a cell nucleus positive staining image;

specifically, the classification of the color value of the cell nucleus image as a positively stained image is a cell nucleus positive stained image, and the classification of the color value of the cell nucleus image as a negatively stained image is a cell nucleus negative stained image. For example, in immunohistochemistry, the nucleus positive staining is dark brown, and the nucleus negative staining is blue, which is not specifically limited by the examples.

S404, classifying according to the color values of the cytoplasm images to obtain a cytoplasm negative staining image and a cytoplasm positive staining image;

specifically, the cytoplasm positive stain image is classified as having a color value of the cytoplasm image being positive stain, and the cytoplasm negative stain image is classified as having a color value of the cytoplasm image being negative stain.

S406, respectively performing color value calculation according to the cell nucleus negative staining image, the cell nucleus positive staining image, the cytoplasm negative staining image and the cytoplasm positive staining image to obtain a cell nucleus negative staining color value, a cell cytoplasm negative staining color value, a cell nucleus positive staining color value and a cell cytoplasm positive staining color value.

Specifically, color values are calculated according to the cell nucleus negative staining image to obtain cell nucleus negative staining color values; performing color value calculation according to the cell nucleus positive staining image to obtain a cell nucleus positive staining color value; performing color value calculation according to the cytoplasm negative staining image to obtain a cytoplasm negative staining color value; and respectively carrying out color value calculation according to the cytoplasm positive staining image to obtain cytoplasm positive staining color values.

The color value calculation can calculate the average value of RGB of all pixel points in the color-dyed image, can also extract the maximum value of RGB of all pixel points in the color-dyed image, and can also extract the minimum value of RGB of all pixel points in the color-dyed image; it is understood that other algorithms may be used for the color value calculation, and the examples herein are not limited in particular.

Optionally, the dyeing color value and the dyeing standard value corresponding to the dyeing color value adopt the same color value calculation method, so that the accuracy and consistency of dyeing quality evaluation are improved.

As shown in fig. 5, in an embodiment, the determining the staining quality of the pathological section corresponding to the medical image to be analyzed according to the plurality of staining color values and the staining standard value includes:

s502, scoring the staining color value and a staining standard value corresponding to the staining color value to obtain a staining color value score;

when the dyeing standard value is a specific RGB numerical value, calculating the difference value between the dyeing color value and the dyeing standard value corresponding to the dyeing color value to obtain a dyeing color difference value, and scoring according to the dyeing color difference value to obtain a dyeing color value score; and when the dyeing standard value is a range value, when the dyeing color value is within the range value of the dyeing standard value corresponding to the dyeing color value, the dyeing color value is scored to be full, and when the dyeing color value is not within the range value of the dyeing standard value corresponding to the dyeing color value, the dyeing color value score is calculated according to a preset scoring rule.

When the dyeing color value is not within the range value of the dyeing standard value corresponding to the dyeing color value, the dyeing color value score is calculated according to a preset scoring rule, and the method comprises the following steps: calculating the average values of the range value starting point and the end point of the range value of the dyeing standard value corresponding to the dyeing color value to obtain the center of the dyeing standard value; when the dyeing color value is smaller than the center of the dyeing standard value, subtracting the dyeing color value from the starting point of the range value of the dyeing standard value corresponding to the dyeing color value to obtain a deviation difference value; and when the dyeing color value is larger than the center of the dyeing standard value, subtracting the range value end point of the range value of the dyeing standard value corresponding to the dyeing color value from the dyeing color value to obtain a deviation difference value, and calculating according to the deviation difference value to obtain a dyeing color value score.

Calculating a staining color value score according to the deviation difference value, wherein the method comprises the following steps: and multiplying the deviation difference value by a preset coefficient to obtain a deviation score, and subtracting the deviation score from the full score to obtain a staining color value score. It is understood that other algorithms may be used to calculate the staining color value score, and the examples are not limited thereto.

S504, carrying out comprehensive scoring according to all the staining color value scores to obtain a pathological section staining comprehensive score;

and weighting and summing all the staining color value scores to obtain a pathological section staining comprehensive score. It is understood that other algorithms may be used for the composite score, and the examples herein are not limited in particular.

Optionally, when all the staining color value scores are weighted and summed, the staining color values may be set with weights according to pathology, the staining color values may also be set with weights according to a staining method, and the staining color values may also be set with weights according to pathology and a staining method, which is not specifically limited in this example.

S506, obtaining a dyeing comprehensive score standard value, and comparing according to the dyeing comprehensive score standard value and the pathological section dyeing comprehensive score to determine the dyeing quality of the pathological section corresponding to the medical image to be analyzed.

Optionally, the standard value of the staining composite score may be a specific RGB value, or may be a range value.

When the staining comprehensive score standard value is a specific RGB numerical value, the staining quality of the pathological section corresponding to the medical image to be analyzed is qualified when the staining comprehensive score of the pathological section is greater than or equal to the staining comprehensive score standard value, and the staining quality of the pathological section corresponding to the medical image to be analyzed is unqualified when the staining comprehensive score of the pathological section is less than the staining comprehensive score standard value; and when the staining comprehensive score standard value is a range value, the staining quality of the pathological section corresponding to the medical image to be analyzed is qualified when the staining comprehensive score of the pathological section is within the range value of the staining comprehensive score standard value, and the staining quality of the pathological section corresponding to the medical image to be analyzed is unqualified when the staining comprehensive score of the pathological section is outside the range value of the staining comprehensive score standard value.

Optionally, the weight of each staining comprehensive scoring standard value may be set according to a pathology, a staining method, or a pathology and a staining method, which is not specifically limited in this example.

According to the embodiment, the accuracy and the objectivity of pathological section staining quality evaluation are improved by establishing an objectivity evaluation system and indexes, the evaluation at each time does not need to depend on professional pathological doctors, the labor cost for evaluating the staining quality is reduced, the subjectivity of different pathological doctors can be prevented from influencing the evaluation accuracy, a uniform evaluation system can be conveniently adopted in the same laboratory and among different laboratories, and the consistency of quality evaluation is ensured.

In one embodiment, the scoring the staining color value and the staining standard value corresponding to the staining color value to obtain a staining color value score includes: calculating the difference value between the dyeing color value and the dyeing standard value corresponding to the dyeing color value to obtain the dyeing color difference value; and scoring according to the dyeing color difference value to obtain a dyeing color value score.

As shown in fig. 6, in one embodiment, the staining color value comprises a channel number of three channels, and the staining standard value comprises a channel standard value of three channels;

calculating the difference value between the dyeing color value and the dyeing standard value corresponding to the dyeing color value to obtain the dyeing color difference value, wherein the method comprises the following steps:

s602, respectively calculating a channel difference value between the channel value of each channel and the channel standard value corresponding to the channel value;

for example, when the color value adopts RGB, the channel value of the R channel of the dyeing color value is subtracted by the channel value of the R channel of the dyeing standard value corresponding to the dyeing color value to obtain an R channel difference value, the channel value of the G channel of the dyeing color value is subtracted by the channel value of the G channel of the dyeing standard value corresponding to the dyeing color value to obtain a G channel difference value, and the channel value of the B channel of the dyeing color value is subtracted by the channel value of the B channel of the dyeing standard value corresponding to the dyeing color value to obtain a B channel difference value, which is not specifically limited in this example.

S604, calculating an absolute value according to the channel difference value to obtain a channel color difference value;

for example, when the color value adopts RGB, an absolute value of the R channel difference is calculated to obtain an R channel color difference, an absolute value of the G channel difference is calculated to obtain a G channel color difference, an absolute value of the B channel difference is calculated to obtain a B channel color difference, and the R channel color difference, the G channel color difference, and the B channel color difference are taken as channel color differences.

And S606, carrying out weighted summation calculation according to the three channel color difference values to obtain a dyeing color difference value.

For example, when the color value adopts RGB, the R channel color difference value, the G channel color difference value, and the B channel color difference value are subjected to weighted summation calculation to obtain a dyeing color difference value.

Optionally, the weight of each channel color difference value may be set according to a pathology, may also be set according to a staining method, and may also be set according to a pathology and a staining method, which is not specifically limited in this example.

In one embodiment, the scoring according to the staining color difference value to obtain a staining color value score includes: and dividing the dyeing color difference value by a preset numerical value to obtain a dyeing color value score. It is understood that other algorithms may be selected, for example, the staining color difference value is directly scored as the staining color value, or the staining color difference value is divided by 255 to obtain a difference index, and 1 is subtracted from the difference index to obtain the staining color value score, which is not limited in this example.

As shown in fig. 7, in an embodiment, a method for evaluating staining quality of a pathological section is further provided, the method comprising:

s702, carrying out digital scanning on the dyed pathological section to obtain an initial medical image;

specifically, the stained pathological section is digitally scanned, resulting in a digitized initial medical image.

S704, carrying out binarization processing on the initial medical image to obtain a binarized medical image;

specifically, an image binarization algorithm is adopted for the initial medical image to obtain a binarization medical image.

Image Binarization (Image Binarization) is a process of setting the gray value of a pixel point on an Image to be 0 or 255, namely, the whole Image presents an obvious black-white effect.

S706, obtaining a color value with the lowest numerical value from the binaryzation medical image to obtain a binaryzation lowest color value;

s708, determining the lowest color value of the initial medical image corresponding to the binarization lowest color value according to the binarization lowest color value;

specifically, the position of the corresponding pixel point is determined according to the binarization lowest color value, and the color value before the pixel point at the position is not binarized is used as the lowest color value of the initial medical image corresponding to the binarization lowest color value.

S710, correcting the initial medical image according to the lowest color value of the initial medical image corresponding to the binarization lowest color value to obtain the medical image to be analyzed;

optionally, a corrected color value is calculated according to the white color value and the lowest color value of the initial medical image corresponding to the binarized lowest color value; and obtaining the medical image to be analyzed according to the initial medical image and the corrected color value.

For example, when the color value adopts RGB, the channel value of the R channel of the lowest color value of the initial medical image corresponding to the binarized lowest color value is subtracted from 255 to obtain the corrected color value of the R channel, the channel value of the G channel of the lowest color value of the initial medical image corresponding to the binarized lowest color value is subtracted from 255 to obtain the corrected color value of the G channel, and the channel value of the B channel of the lowest color value of the initial medical image corresponding to the binarized lowest color value is subtracted from 255 to obtain the corrected color value of the B channel; subtracting the corrected color value of the R channel from the channel value of the R channel of all pixel points of the initial medical image, subtracting the corrected color value of the G channel from the channel value of the G channel of all pixel points of the initial medical image, subtracting the corrected color value of the B channel from the channel value of the B channel of all pixel points of the initial medical image, and taking the subtracted initial medical image as the medical image to be analyzed, which is not specifically limited in this example.

The lowest color value of the initial medical image corresponding to the binarized lowest color value is referred to as a background color, which is normally white, and the RGB color value of white is (255,255,255), and the background color changes to other colors when illumination is different in the digital scanning.

S712, performing feature recognition on the medical image to be analyzed to obtain a plurality of dyeing color values;

s714, obtaining a dyeing standard value, and determining the dyeing quality of the pathological section corresponding to the medical image to be analyzed according to the plurality of dyeing color values and the dyeing standard value.

In the embodiment, the initial medical image is corrected to obtain the medical image to be analyzed, and then the dyeing quality of the medical image to be analyzed is evaluated, so that chromatic aberration caused by illumination in digital scanning is avoided, and the accuracy of pathological section dyeing quality evaluation is further improved.

In one embodiment, after the determining the staining quality of the pathological section corresponding to the medical image to be analyzed according to the plurality of staining color values and the staining standard value, the method further includes: acquiring an evaluation information table, wherein the evaluation information table comprises laboratory identification, stainer identification and dyeing process data corresponding to the dyed pathological section; and updating the evaluation information table, the medical image to be analyzed, the plurality of dyeing color values, the dyeing standard value and the dyeing quality of the pathological section corresponding to the medical image to be analyzed to a pathological section dyeing database. Specifically, an evaluation information table input by a user is acquired, and then the evaluation information table corresponding to a stained pathological section, the medical image to be analyzed, the plurality of staining color values, the staining standard value, and the staining quality of the pathological section corresponding to the medical image to be analyzed are added to the pathological section staining database.

The user may be a person who dyes a pathological section, and this example is not limited in particular.

The staining process data includes reagent adding data, antibody adding data, and incubation conditions, the reagent adding data includes reagent name, time, and measurement, and the antibody adding data includes antibody name, time, and measurement, which is not limited specifically here.

The stainer identifier is used to uniquely identify a stainer, and may be a name, an ID, or the like, which is not specifically limited herein.

As shown in fig. 8, in one embodiment, the method further comprises:

s802, acquiring the laboratory mark;

the laboratory identifier is used for uniquely identifying a laboratory, and may be a laboratory name or a laboratory ID number, which is not specifically limited herein.

S804, acquiring the evaluation information table, the medical image to be analyzed, the dyeing color values and the dyeing standard value corresponding to unqualified dyeing quality of the pathological section from a pathological section dyeing database according to the laboratory identification;

s806, analyzing unqualified reasons according to the evaluation information table, the medical image to be analyzed, the dyeing color values and the dyeing standard value to obtain unqualified reason data.

The analysis of the failure cause includes a reagent addition amount, a reagent addition time, a reagent addition sequence, an antibody addition amount, an antibody addition time, and an antibody addition sequence, which are not specifically limited by the examples herein.

In one embodiment, after the analyzing the unqualified reason according to the evaluation information table, the medical image to be analyzed, the plurality of dyeing color values and the dyeing standard value to obtain the unqualified reason data, the method further comprises: acquiring an improvement suggestion list; and obtaining improvement suggestions from the improvement suggestion list according to the unqualified reason data to obtain a laboratory improvement suggestion list. Specifically, the unqualified reason data is searched from the improvement suggestion list, and the improvement suggestion list is obtained through matching.

The improvement suggestion list is a reconstruction library established after an expert pathologist comprehensively evaluates the multiple stained pathological sections.

According to the embodiment, the laboratory improvement suggestion list is obtained through data analysis, and the laboratory is facilitated to improve the operation specification, the inspection specification and the training system of the laboratory according to the laboratory improvement suggestion list.

As shown in fig. 9, in one embodiment, there is also provided a pathological section staining quality evaluation device, including:

a scanning module 902, configured to perform digital scanning on the stained pathological section to obtain a medical image to be analyzed;

a staining color value extracting module 904, configured to perform feature recognition on the medical image to be analyzed to obtain a plurality of staining color values;

and a staining quality determining module 906, configured to obtain a staining standard value, and determine, according to the plurality of staining color values and the staining standard value, a staining quality of a pathological section corresponding to the medical image to be analyzed.

The pathological section staining quality evaluation device of the embodiment digitally scans the stained pathological sections to obtain medical images to be analyzed, and analyzes the digitized medical images to be analyzed through artificial intelligence, so that the subjective evaluation of pathologists or pathological technicians is avoided, and the consistency of quality evaluation is ensured; treat the medical image of analysis and carry out feature recognition, obtain a plurality of dyeing colour values, acquire the dyeing standard value, confirm the dyeing quality with the pathological section that treats analysis medical image and correspond according to a plurality of dyeing colour values and dyeing standard value, the method of discerning the dyeing colour value is the same, adopt a plurality of dyeing colour values and dyeing standard value to evaluate, the evaluation parameter is unified, the evaluation standard is unified, adopt artificial intelligence, thereby be fit for pathological section dyeing quality evaluation in the same laboratory, also be fit for pathological section dyeing quality evaluation between different laboratories.

In one embodiment, the scanning module 902 includes a digital scanning sub-module, a color correction sub-module;

the digital scanning sub-module is used for digitally scanning the dyed pathological section to obtain an initial medical image;

the color correction submodule is used for carrying out binarization processing on the initial medical image to obtain a binarization medical image, obtaining a color value with the lowest numerical value from the binarization medical image to obtain a binarization minimum color value, determining the minimum color value of the initial medical image corresponding to the binarization minimum color value according to the binarization minimum color value, and carrying out correction processing on the initial medical image according to the minimum color value of the initial medical image corresponding to the binarization minimum color value to obtain the medical image to be analyzed.

In one embodiment, the apparatus further comprises:

and the data updating module is used for acquiring an evaluation information table which comprises laboratory identification, stainer identification and dyeing process data corresponding to the dyed pathological section, and updating the evaluation information table, the medical image to be analyzed, the dyeing color values, the dyeing standard value and the dyeing quality of the pathological section corresponding to the medical image to be analyzed to a pathological section dyeing database.

In one embodiment, the apparatus further comprises:

and the laboratory quality control module is used for acquiring the laboratory identification, acquiring the evaluation information table, the medical image to be analyzed, the dyeing color values and the dyeing standard value corresponding to the unqualified dyeing quality of the pathological section from a pathological section dyeing database according to the laboratory identification, and analyzing the unqualified reasons according to the evaluation information table, the medical image to be analyzed, the dyeing color values and the dyeing standard value to obtain unqualified reason data.

FIG. 10 is a diagram illustrating an internal structure of a computer device in one embodiment. The computer device may specifically be a terminal, and may also be a server. As shown in fig. 10, the computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program which, when executed by the processor, causes the processor to implement the method for assessing the staining quality of a pathological section. The internal memory may also store a computer program, and the computer program, when executed by the processor, may cause the processor to perform a method of assessing staining quality of a pathological section. Those skilled in the art will appreciate that the architecture shown in fig. 10 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, a method for evaluating staining quality of a pathological section provided by the present application can be implemented in the form of a computer program, and the computer program can be run on a computer device as shown in fig. 10. The memory of the computer device can store various program templates of the pathological section staining quality evaluation device. Such as a scanning module 902, a stain color value extraction module 904, and a stain quality determination module 906.

In one embodiment, the present invention also proposes a storage medium storing a computer program of instructions which, when executed by a processor, causes the processor to carry out the following method steps when executed:

The method comprises the steps of digitally scanning the dyed pathological section to obtain a medical image to be analyzed, and analyzing the digital medical image to be analyzed through artificial intelligence, so that subjective evaluation of a pathologist or a pathology technician is avoided, and consistency of quality evaluation is ensured; treat the medical image of analysis and carry out feature recognition, obtain a plurality of dyeing colour values, acquire the dyeing standard value, confirm the dyeing quality with the pathological section that treats analysis medical image and correspond according to a plurality of dyeing colour values and dyeing standard value, the method of discerning the dyeing colour value is the same, adopt a plurality of dyeing colour values and dyeing standard value to evaluate, the evaluation parameter is unified, the evaluation standard is unified, adopt artificial intelligence, thereby be fit for pathological section dyeing quality evaluation in the same laboratory, also be fit for pathological section dyeing quality evaluation between different laboratories.

In one embodiment, the feature recognition of the medical image to be analyzed to obtain a plurality of staining color values includes: identifying the cell structure characteristics according to the medical image to be analyzed to obtain a cell nucleus image and a cell cytoplasm image; and respectively calculating color values according to the cell nucleus image and the cell pulp image to obtain a plurality of staining color values, wherein the plurality of staining color values comprise cell nucleus negative staining color values, cell pulp negative staining color values, cell nucleus positive staining color values and/or cell pulp positive staining color values.

In one embodiment, the performing cell structure feature recognition according to the medical image to be analyzed to obtain a cell nucleus image and a cell cytoplasm image includes: segmenting cell nucleuses according to the medical image to be analyzed by adopting a cell nucleus segmentation algorithm to obtain a cell nucleus image; and segmenting cytoplasm by adopting a cytoplasm segmentation algorithm according to the medical image to be analyzed to obtain a cytoplasm image.

In one embodiment, the performing color value calculation according to the cell nucleus image and the cell cytoplasm image respectively to obtain a plurality of staining color values, where the plurality of staining color values include a cell nucleus negative staining color value, a cell cytoplasm negative staining color value, a cell nucleus positive staining color value, and/or a cell cytoplasm positive staining color value, includes: classifying according to the color values of the cell nucleus images to obtain a cell nucleus negative staining image and a cell nucleus positive staining image; classifying according to the color values of the cytoplasm images to obtain a cytoplasm negative staining image and a cytoplasm positive staining image; and respectively carrying out color value calculation according to the cell nucleus negative staining image, the cell nucleus positive staining image, the cytoplasm negative staining image and the cytoplasm positive staining image to obtain a cell nucleus negative staining color value, a cell cytoplasm negative staining color value, a cell nucleus positive staining color value and a cell cytoplasm positive staining color value.

In one embodiment, the staining standard values comprise: a standard value for cell nucleus negative staining, a standard value for cell cytoplasm negative staining, a standard value for cell nucleus positive staining and a standard value for cell cytoplasm positive staining.

In one embodiment, the determining the staining quality of the pathological section corresponding to the medical image to be analyzed according to the plurality of staining color values and the staining standard value comprises: scoring the staining color value and a staining standard value corresponding to the staining color value to obtain a staining color value score; performing comprehensive scoring according to all the staining color value scores to obtain a pathological section staining comprehensive score; and acquiring a dyeing comprehensive score standard value, and comparing and determining the dyeing quality of the pathological section corresponding to the medical image to be analyzed according to the dyeing comprehensive score standard value and the pathological section dyeing comprehensive score.

In one embodiment, the staining color value comprises a channel number of three channels, and the staining standard value comprises a channel standard value of three channels; calculating the difference value between the dyeing color value and the dyeing standard value corresponding to the dyeing color value to obtain the dyeing color difference value, wherein the method comprises the following steps: respectively calculating a channel difference value between the channel value of each channel and the channel standard value corresponding to the channel value; calculating an absolute value according to the channel difference value to obtain a channel color difference value; and performing weighted summation calculation according to the three channel color difference values to obtain a dyeing color difference value.

In one embodiment, the scoring according to the staining color difference value to obtain a staining color value score includes: and dividing the dyeing color difference value by a preset numerical value to obtain a dyeing color value score.

In one embodiment, the digitally scanning the stained pathological section to obtain the medical image to be analyzed includes: carrying out digital scanning on the dyed pathological section to obtain an initial medical image; carrying out binarization processing on the initial medical image to obtain a binarized medical image; obtaining a color value with the lowest numerical value from the binaryzation medical image to obtain a binaryzation lowest color value; determining a lowest color value of the initial medical image corresponding to the binarization lowest color value according to the binarization lowest color value; and correcting the initial medical image according to the lowest color value of the initial medical image corresponding to the binarization lowest color value to obtain the medical image to be analyzed.

In an embodiment, the modifying the initial medical image according to the lowest color value of the initial medical image corresponding to the binarized lowest color value to obtain the medical image to be analyzed includes:

calculating a corrected color value according to the white color value and the lowest color value of the initial medical image corresponding to the binarization lowest color value; and obtaining the medical image to be analyzed according to the initial medical image and the corrected color value.

In one embodiment, after the determining the staining quality of the pathological section corresponding to the medical image to be analyzed according to the plurality of staining color values and the staining standard value, the method further includes: acquiring an evaluation information table, wherein the evaluation information table comprises laboratory identification, stainer identification and dyeing process data corresponding to the dyed pathological section; and updating the evaluation information table, the medical image to be analyzed, the plurality of dyeing color values, the dyeing standard value and the dyeing quality of the pathological section corresponding to the medical image to be analyzed to a pathological section dyeing database.

In one embodiment, the method further comprises: acquiring the laboratory identification; acquiring the evaluation information table, the medical image to be analyzed, the dyeing color values and the dyeing standard value corresponding to unqualified dyeing quality of the pathological section from a pathological section dyeing database according to the laboratory identification; and analyzing unqualified reasons according to the evaluation information table, the medical image to be analyzed, the plurality of dyeing color values and the dyeing standard value to obtain unqualified reason data.

In one embodiment, after the analyzing the unqualified reason according to the evaluation information table, the medical image to be analyzed, the plurality of dyeing color values and the dyeing standard value to obtain the unqualified reason data, the method further comprises: acquiring an improvement suggestion list; and obtaining improvement suggestions from the improvement suggestion list according to the unqualified reason data to obtain a laboratory improvement suggestion list.

In one embodiment, the present invention also proposes a computer device comprising at least one memory storing a computer program of instructions, at least one processor, the computer program of instructions, when executed by the processor, causing the processor to perform the method steps of:

It should be noted that, the pathological section staining quality evaluation method, the pathological section staining quality evaluation device, the storage medium and the computer device described above belong to a general inventive concept, and the contents in the embodiments of the pathological section staining quality evaluation method, the pathological section staining quality evaluation device, the storage medium and the computer device are mutually applicable.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a non-volatile computer-readable storage medium, and can include the processes of the embodiments of the methods described above when the program is executed. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A pathological section staining quality evaluation method comprises the following steps:

2. The method for evaluating staining quality of pathological sections according to claim 1, wherein the step of performing feature recognition on the medical image to be analyzed to obtain a plurality of staining color values comprises:

identifying the cell structure characteristics according to the medical image to be analyzed to obtain a cell nucleus image and a cell cytoplasm image;

and respectively calculating color values according to the cell nucleus image and the cell pulp image to obtain a plurality of staining color values, wherein the plurality of staining color values comprise cell nucleus negative staining color values, cell pulp negative staining color values, cell nucleus positive staining color values and/or cell pulp positive staining color values.

3. The method for evaluating the staining quality of pathological sections according to claim 2, wherein the identifying the cell structural features according to the medical image to be analyzed to obtain the cell nucleus image and the cell cytoplasm image comprises:

segmenting cell nucleuses according to the medical image to be analyzed by adopting a cell nucleus segmentation algorithm to obtain a cell nucleus image;

and segmenting cytoplasm by adopting a cytoplasm segmentation algorithm according to the medical image to be analyzed to obtain a cytoplasm image.

4. The method of claim 2, wherein the calculating color values according to the nuclear image and the cytoplasm image to obtain a plurality of staining color values, wherein the staining color values include nuclear negative staining color values, cytoplasm negative staining color values, nuclear positive staining color values and/or cytoplasm positive staining color values, comprises:

classifying according to the color values of the cell nucleus images to obtain a cell nucleus negative staining image and a cell nucleus positive staining image;

classifying according to the color values of the cytoplasm images to obtain a cytoplasm negative staining image and a cytoplasm positive staining image;

and respectively carrying out color value calculation according to the cell nucleus negative staining image, the cell nucleus positive staining image, the cytoplasm negative staining image and the cytoplasm positive staining image to obtain a cell nucleus negative staining color value, a cell cytoplasm negative staining color value, a cell nucleus positive staining color value and a cell cytoplasm positive staining color value.

5. The method for evaluating the staining quality of a pathological section according to claim 1, wherein the staining standard value comprises: a standard value for cell nucleus negative staining, a standard value for cell cytoplasm negative staining, a standard value for cell nucleus positive staining and a standard value for cell cytoplasm positive staining.

6. The method for evaluating staining quality of a pathological section according to claim 1, wherein the determining the staining quality of the pathological section corresponding to the medical image to be analyzed according to the plurality of staining color values and the staining standard value comprises:

scoring the staining color value and a staining standard value corresponding to the staining color value to obtain a staining color value score;

performing comprehensive scoring according to all the staining color value scores to obtain a pathological section staining comprehensive score;

and acquiring a dyeing comprehensive score standard value, and comparing and determining the dyeing quality of the pathological section corresponding to the medical image to be analyzed according to the dyeing comprehensive score standard value and the pathological section dyeing comprehensive score.

7. The method for evaluating staining quality of a pathological section according to claim 6, wherein the scoring the staining color value and a staining standard value corresponding to the staining color value to obtain a staining color value score comprises:

calculating the difference value between the dyeing color value and the dyeing standard value corresponding to the dyeing color value to obtain the dyeing color difference value;

and scoring according to the dyeing color difference value to obtain a dyeing color value score.

8. The method for evaluating the staining quality of a pathological section according to claim 7, wherein the staining color values comprise channel numerical values of three channels, and the staining standard values comprise channel standard values of three channels;

respectively calculating a channel difference value between the channel value of each channel and the channel standard value corresponding to the channel value;

calculating an absolute value according to the channel difference value to obtain a channel color difference value;

and performing weighted summation calculation according to the three channel color difference values to obtain a dyeing color difference value.

9. The method for evaluating staining quality of pathological sections according to claim 7, wherein the scoring according to the staining color difference value to obtain a staining color value score comprises:

and dividing the dyeing color difference value by a preset numerical value to obtain a dyeing color value score.

10. The method for evaluating the staining quality of the pathological section according to any one of claims 1 to 9, wherein the digital scanning of the stained pathological section to obtain the medical image to be analyzed comprises:

carrying out digital scanning on the dyed pathological section to obtain an initial medical image;

carrying out binarization processing on the initial medical image to obtain a binarized medical image;

obtaining a color value with the lowest numerical value from the binaryzation medical image to obtain a binaryzation lowest color value;

determining a lowest color value of the initial medical image corresponding to the binarization lowest color value according to the binarization lowest color value;

and correcting the initial medical image according to the lowest color value of the initial medical image corresponding to the binarization lowest color value to obtain the medical image to be analyzed.

11. The pathological section staining quality evaluation method according to claim 10, wherein the modifying the initial medical image according to the lowest color value of the initial medical image corresponding to the binarized lowest color value to obtain the medical image to be analyzed comprises:

12. The method for evaluating staining quality of a pathological section according to any one of claims 1 to 9, further comprising, after the determining the staining quality of the pathological section corresponding to the medical image to be analyzed from the plurality of staining color values and the staining standard value:

acquiring an evaluation information table, wherein the evaluation information table comprises laboratory identification, stainer identification and dyeing process data corresponding to the dyed pathological section;

and updating the evaluation information table, the medical image to be analyzed, the plurality of dyeing color values, the dyeing standard value and the dyeing quality of the pathological section corresponding to the medical image to be analyzed to a pathological section dyeing database.

13. The method for evaluating staining quality of pathological sections according to claim 12, further comprising:

acquiring the laboratory identification;

acquiring the evaluation information table, the medical image to be analyzed, the dyeing color values and the dyeing standard value corresponding to unqualified dyeing quality of the pathological section from a pathological section dyeing database according to the laboratory identification;

and analyzing unqualified reasons according to the evaluation information table, the medical image to be analyzed, the plurality of dyeing color values and the dyeing standard value to obtain unqualified reason data.

14. The method for evaluating staining quality of pathological sections according to claim 13, further comprising, after the analyzing the cause of failure according to the evaluation information table, the medical image to be analyzed, the plurality of staining color values, and the staining standard value to obtain the cause of failure data:

acquiring an improvement suggestion list;

and obtaining improvement suggestions from the improvement suggestion list according to the unqualified reason data to obtain a laboratory improvement suggestion list.

15. A pathological section staining quality evaluation device, characterized in that, the device includes:

16. A storage medium storing a computer program of instructions which, when executed by a processor, causes the processor to carry out the steps of the method according to any one of claims 1 to 14.

17. A computer device comprising at least one memory storing a program of computer instructions which, when executed by the processor, causes the processor to perform the steps of the method of any one of claims 1 to 14, at least one processor.