Method, device and equipment for evaluating dry syndrome labial gland pathological image and storage medium
By dividing and correcting the weight coefficients of labial gland pathology images, and combining them with a deep learning model, the problems of high cost and low accuracy of manual interpretation are solved, and more accurate assessment of lymphocyte foci is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- UNIV OF ELECTRONICS SCI & TECH OF CHINA
- Filing Date
- 2026-03-10
- Publication Date
- 2026-05-29
AI Technical Summary
In the current technology, the pathological diagnosis of labial gland biopsy sections relies on manual observation, which is costly and inaccurate, and makes it difficult to accurately assess the lymphocytic foci grading of Sjögren's syndrome.
By dividing the pathological images of the labial glands into regions, determining the weight coefficients based on the pathological microenvironment characteristics of the sub-regions, performing weighted summation, and combining the heterogeneity correction coefficient to correct the lymphocyte foci score, the deep learning model is used to extract features and perform precise transformation.
It improves the accuracy of lymphocyte foci scoring, captures pathological details and lesion distribution, avoids missed diagnoses in traditional methods, and provides more accurate lesion assessment.
Smart Images
Figure CN122115401A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of image processing technology. Specifically, it relates to methods, apparatus, devices, and storage media for evaluating pathological images of labial glands in Sjögren's syndrome. Background Technology
[0002] Sjögren's syndrome (SS) is a complex, chronic, systemic autoimmune disease characterized by hypofunction of exocrine glands (particularly salivary and lacrimal glands) due to lymphocytic infiltration. Its clinical diagnostic criteria typically require a combination of clinical manifestations, serological markers, and histopathological examination results.
[0003] Labial gland biopsy is currently recognized internationally as one of the key and highly specific diagnostic methods for confirming SS. Pathological examination mainly assesses the degree of lymphocyte aggregation in the labial gland tissue, i.e., lymphocyte foci grading (such as according to the Focus Score (FS)), to determine whether it meets the pathological diagnostic criteria for SS.
[0004] Currently, the pathological diagnosis of labial gland biopsy sections mainly relies on manual observation and counting by pathologists under an optical microscope. This traditional manual interpretation method is costly and inaccurate. Summary of the Invention
[0005] This disclosure provides a method, apparatus, device, and storage medium for evaluating pathological images of labial glands in Sjögren's syndrome.
[0006] According to one aspect of this disclosure, a method for evaluating pathological images of labial glands in Sjögren's syndrome is provided, comprising: The labial gland tissue region in the pathological image of the labial gland of the subject to be examined is divided into multiple sub-regions; Based on the pathological microenvironment characteristics of each sub-region, the weight coefficients of each sub-region are determined respectively; Based on the weight coefficients of each sub-region, the ratio of the number of lymphoid foci to the area of each sub-region is weighted and summed to obtain the lymphocyte foci score of the subject's dry mouth syndrome. Based on the distribution of lymph nodes in the labial gland tissue region across the various image sub-regions, a heterogeneity correction coefficient is determined. Based on the heterogeneity correction coefficient, the lymphocyte foci score of the subject's dry mouth syndrome is corrected.
[0007] According to one aspect of this disclosure, an evaluation device for pathological images of labial glands in Sjögren's syndrome is provided, comprising: The region segmentation module is used to divide the labial gland tissue region in the labial gland pathology image of the subject to be examined into multiple sub-regions; The weight coefficient determination module is used to determine the weight coefficient of each sub-region based on the pathological microenvironment characteristics of each sub-region. The scoring determination module is used to perform a weighted summation of the ratio of the number of lymphoid foci to the area of each sub-region based on the weight coefficient of each sub-region, so as to obtain the lymphoid foci score of the subject's dry mouth syndrome. The correction coefficient determination module is used to determine the heterogeneity correction coefficient based on the distribution of lymph nodes in the labial gland tissue region across each of the image sub-regions. The scoring correction module is used to correct the lymphocyte foci score of the subject's dry mouth syndrome based on the heterogeneity correction coefficient.
[0008] According to another aspect of this disclosure, an electronic device is provided, comprising: At least one processor; and The memory is communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform any of the methods for evaluating Sjögren's syndrome labial gland pathology images in the embodiments of this disclosure.
[0009] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions for causing the computer to perform an evaluation method for any Sjögren's syndrome labial gland pathology image according to any embodiment of this disclosure.
[0010] According to the technique disclosed herein, the labial gland tissue region in the pathological image of the labial gland of the subject to be examined is divided into multiple sub-regions; based on the pathological microenvironment characteristics of each sub-region, a weight coefficient for each sub-region is determined; based on the weight coefficients of each sub-region, the ratio of the number of lymphoid foci to the area of each sub-region is weighted and summed to obtain the lymphocyte foci score of the subject's dry labial gland syndrome; based on the distribution of lymphoid foci in each image sub-region of the labial gland tissue region, a heterogeneity correction coefficient is determined; based on the heterogeneity correction coefficient, the lymphocyte foci score of the subject's dry labial gland syndrome is corrected. On the one hand, the pathological microenvironment characteristics of each sub-region are used to determine the weight coefficient of each region; on the other hand, the heterogeneity of lesion distribution in glandular tissue is considered, thus improving the accuracy of the lymphocyte foci score for dry labial gland syndrome.
[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description
[0012] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein: Figure 1 This is a flowchart of a method for evaluating pathological images of labial glands in Sjögren's syndrome according to an embodiment of this disclosure; Figure 2 This is a flowchart of a method for evaluating pathological images of labial glands in Sjögren's syndrome according to another embodiment of this disclosure; Figure 3 This is a structural block diagram of an evaluation device for labial gland pathology images in Sjögren's syndrome according to an embodiment of the present disclosure; Figure 4 This is a block diagram of an electronic device according to an embodiment of the present disclosure. Detailed Implementation
[0013] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.
[0014] Figure 1 This is a flowchart of a method for evaluating pathological images of labial glands in Sjögren's syndrome, as disclosed in this article.
[0015] like Figure 1 As shown, the evaluation method for the pathological images of the labial glands in Sjögren's syndrome may include: S110, the labial gland tissue region in the labial gland pathology image of the subject to be examined is divided into multiple sub-regions; S120, based on the pathological microenvironment characteristics of each sub-region, determines the weight coefficient of each sub-region respectively; S130, based on the weight coefficients of each sub-region, the ratio of the number of lymphoid foci to the area of each sub-region is weighted and summed to obtain the lymphocyte foci score of the subject's dry labial syndrome. S140, Based on the distribution of lymph nodes in the labial gland tissue region across various image sub-regions, determine the heterogeneity correction coefficient; S150, based on the heterogeneity correction coefficient, corrects the lymphocyte foci score of the subject's dry mouth syndrome.
[0016] For example, a digital pathology scanner is used to perform whole-slide scanning on stained glass slides of labial gland tissue to generate a high-resolution whole-slide image (WSI), which is a pathological image of the labial gland of the subject being examined.
[0017] For example, a spatial gridding method is used to divide the labial gland tissue region in the pathological image of the labial gland of the subject to be examined into multiple sub-regions. Each sub-region is a region image block.
[0018] For example, the pathological microenvironment features of a sub-region image patch include the lymphocyte density, germinal center structures, and distances to adjacent ducts. Germinal center structures can be understood as, for example, whether the sub-region image patch includes germinal center structures. These features can be obtained by feature extraction for each sub-region image patch using a deep learning model.
[0019] For example, after determining the weight coefficients of each sub-region by utilizing the pathological microenvironment characteristics of each sub-region, the weight coefficients of each sub-region are normalized so that their values are all in the range of 0-1.
[0020] For example, normalized weighting coefficients are used to weight and sum the ratios of the number of lymphoid foci to the area of each sub-region to obtain the lymphocytic foci score for the subject's dry mouth syndrome. The number of lymphoid foci and the area of each sub-region can be learned using a deep learning model.
[0021] For example, the lymphocyte foci score of the subject's dry mouth syndrome is corrected based on the heterogeneity correction coefficient and the coefficient value corresponding to the interval in which the heterogeneity correction coefficient falls. For instance, if the heterogeneity correction coefficient falls in the first interval, the coefficient of the first interval is multiplied by the heterogeneity correction coefficient, and the product is multiplied by the lymphocyte foci score of the subject's dry mouth syndrome to obtain the corrected lymphocyte foci score of the subject's dry mouth syndrome.
[0022] According to the above implementation method, each sub-region within the labial gland tissue region is assigned a corresponding weight coefficient based on its pathological microenvironment characteristics, and the lymphocyte foci score of the labial gland tissue region is calculated using this weighted average. Furthermore, a heterogeneity correction coefficient determined based on the lesion distribution across the entire region is introduced to correct the score. In this way, the score can accurately capture pathological details and lesion distribution that are easily overlooked in manual interpretation, improving the accuracy of the score.
[0023] In one implementation, the weight coefficients of each sub-region are determined based on their respective pathological microenvironment characteristics, including: determining the weight coefficients of the sub-regions based on the lymphocyte density, germinal center structures, and distance to adjacent ducts; wherein, the lymphocyte density of the sub-region is positively correlated with the weight coefficient of the sub-region, whether the sub-region includes germinal center structures is positively or negatively correlated with the value of the weight coefficient of the sub-region, and the distance to adjacent ducts of the sub-region is negatively correlated with the weight coefficient of the sub-region.
[0024] In one implementation, the lymphocyte foci score of the subject's dry mouth syndrome is corrected based on a heterogeneity correction coefficient, including: determining the corrected lymphocyte foci score of the subject's dry mouth syndrome based on the product between the heterogeneity correction coefficient and the subject's lymphocyte foci score.
[0025] In this example, a distributional heterogeneity factor (heterogeneity correction coefficient γ) is introduced to nonlinearly correct the final score based on the dispersion of lesions in the whole slice (based on standard deviation or information entropy algorithms). This correction aims to accurately assess complex cases with highly uneven lesion distribution, avoiding missed diagnoses caused by "averaging" in traditional methods. Furthermore, a standardized transformation is performed: the system automatically performs a precise conversion from pixel-level to physical area (mm²) and combines the results with the aforementioned spatial model. Output: WFS value (weighted focus score) accurate to two decimal places. This index is not only used for the preliminary diagnostic classification of Sjögren's syndrome (SS), but also serves as a core quantitative indicator for assessing the spatial extent of disease involvement.
[0026] In one embodiment, the labial gland tissue region in the labial gland pathology image of the subject to be examined is divided into multiple sub-regions, including: separating the blank background and the labial gland tissue region in the labial gland pathology image of the subject to be examined to obtain the labial gland tissue region in the labial gland pathology image; cropping the labial gland tissue region using a sliding window to obtain multiple sub-regions; and using a color transfer algorithm to convert the multiple sub-regions to the same color space to obtain multiple sub-regions after eliminating staining differences.
[0027] For example, tissue region segmentation (ROI extraction): The WSI image is scaled and the background is removed. Using Otsu thresholding or morphological processing algorithms, the region containing labial gland tissue is separated from the blank background to generate a tissue mask.
[0028] For example, image patching: Given the large size of WSI images, the extracted tissue regions are cropped into several local image patches using a sliding window of a preset size (512×512 pixels).
[0029] For example, color normalization: A color transfer algorithm (Reinhard method) is used to convert all image patches to a uniform color space to eliminate the impact of differences in color depth between different batches on the model.
[0030] In this example, preprocessing the image can improve the accuracy of subsequent image feature extraction.
[0031] In one embodiment, the method further includes: determining the acute inflammation index of the subject's dry mouth syndrome based on the inflammatory cell density and inflammatory area of the labial gland tissue region; determining the chronic damage index of the subject's dry mouth syndrome based on the proportion of glandular atrophy area in the labial gland tissue region to the total glandular area and the degree of sclerosis of the ducts in the labial gland tissue region; and determining the comprehensive assessment result of the subject's dry mouth syndrome based on the modified lymphocyte foci score, acute inflammation index, and chronic damage index.
[0032] For example, a weighted formula or regression model can be used to process the inflammatory cell density and inflammatory area of the labial gland tissue region to obtain the acute inflammation index of the subject's dry mouth syndrome.
[0033] For example, the higher the acute inflammation index of dry mouth syndrome, the higher the inflammatory activity of dry mouth syndrome in the subject being tested.
[0034] For example, a weighted formula or regression model can be used to process the proportion of glandular atrophy area in the labial gland tissue region to the total glandular area, as well as the degree of hardening of ducts in the labial gland tissue region, to obtain the chronic damage index of dry labial gland syndrome in the subject to be tested.
[0035] For example, the Chronicity Injury Index is used to assess the degree of irreversible structural damage in a disease and is a key reference for judging the course of the disease and prognosis.
[0036] For example, a visual interactive terminal can be used to display the original pathological images of the subject under examination, some lymphocyte foci and glandular regions of the images, as well as the corrected lymphocyte foci score, acute inflammation index and chronic damage index for dry labial gland syndrome.
[0037] In one embodiment, the method further includes: inputting the labial gland tissue region and each sub-region into a shared encoder to obtain image features of the labial gland tissue region and each sub-region output by the shared encoder; inputting the image features of each sub-region into a first decoder to obtain the number of lymph nodes and the area of each sub-region output by the first decoder; inputting the image features of the labial gland tissue region into a second decoder to obtain the inflammatory cell density and inflammatory area of the labial gland tissue region output by the second decoder; and inputting the image features of the labial gland tissue region into a third decoder to obtain the glandular atrophy area and total glandular area of the labial gland tissue region output by the third decoder.
[0038] like Figure 2 As shown, the neural network model in this example employs a network structure with a shared encoder and three decoders. The first decoder handles the following tasks: Glandular tissue and lymphoid foci identification (focus scoring basis): accurately segmenting effective glandular tissue regions (for calculating A); identifying and counting effective lymphoid foci (for calculating N). Output: Total effective glandular area (A) and total number of effective foci (N). The second decoder handles the following tasks: Acute inflammatory cell and infiltration pattern identification (activity assessment): accurately identifying and quantifying inflammatory cells reflecting disease activity, primarily neutrophils and plasma cells. Simultaneously, it identifies and quantifies diffuse infiltration patterns and focal aggregation patterns. Output: Cell density and infiltration area data required to calculate the Acute Inflammation Index (AII). The third decoder handles the following tasks: Chronic structural damage identification and quantification (chronicity assessment): accurately identifying and segmenting pathological structures representing irreversible damage, primarily including: areas of glandular cell atrophy, ductal sclerosis or fibrosis, and interstitial fibrosis. Output: Glandular atrophy rate and fibrosis area required to calculate the Chronic Damage Index (CDI).
[0039] Figure 3 This is an evaluation device for pathological images of labial glands in Sjögren's syndrome according to an embodiment of the present disclosure.
[0040] like Figure 3 As shown, the evaluation device for the pathological images of the labial glands in Sjögren's syndrome includes: The region division module 310 is used to divide the labial gland tissue region in the labial gland pathology image of the subject to be examined into multiple sub-regions; The weight coefficient determination module 320 is used to determine the weight coefficient of each sub-region based on the pathological microenvironment characteristics of each sub-region. The scoring determination module 330 is used to perform a weighted summation of the ratio of the number of lymphoid foci to the area of each sub-region based on the weight coefficient of each sub-region, so as to obtain the lymphoid foci score of the subject's dry mouth syndrome. The correction coefficient determination module 340 is used to determine the heterogeneity correction coefficient based on the distribution of lymph nodes in the labial gland tissue region in each of the image sub-regions; The scoring correction module 350 is used to correct the lymphocyte foci score of the subject's dry mouth syndrome based on the heterogeneity correction coefficient.
[0041] In one implementation, the weighting coefficient determination module is specifically used for: Based on the lymphocyte density, germinal center structure, and distance to adjacent ducts of the sub-region, a weight coefficient for the sub-region is determined; wherein, the lymphocyte density of the sub-region is positively correlated with the weight coefficient of the sub-region, whether the sub-region includes the germinal center structure is positively or negatively correlated with the value of the weight coefficient of the sub-region, and the distance to adjacent ducts of the sub-region is negatively correlated with the weight coefficient of the sub-region.
[0042] In one embodiment, the scoring correction module 350 is specifically used for: The corrected lymphocyte score for dry mouth syndrome of the subject is determined by multiplying the heterogeneity correction coefficient with the lymphocyte score of dry mouth syndrome of the subject.
[0043] In one embodiment, the region division module 310 includes: The segmentation unit is used to separate the blank background and the labial gland tissue region in the labial gland pathology image of the subject to be examined, so as to obtain the labial gland tissue region in the labial gland pathology image. The cropping unit is used to crop the labial gland tissue region using a sliding window to obtain multiple sub-regions; The color normalization unit is used to convert the multiple sub-regions to the same color space using a color transfer algorithm, so as to obtain the multiple sub-regions after eliminating the color differences.
[0044] In one implementation, it further includes: An acute inflammation index determination module is used to determine the acute inflammation index of dry mouth syndrome in the subject under test based on the inflammatory cell density and inflammatory area of the labial gland tissue region. The chronic damage index determination module is used to determine the chronic damage index of dry mouth syndrome of the subject under test based on the proportion of glandular atrophy area in the labial gland tissue region to the total glandular area and the degree of hardening of ducts in the labial gland tissue region. The assessment result determination module is used to determine the comprehensive assessment result of the subject's dry mouth syndrome based on the modified lymphocyte foci score, acute inflammation index, and chronic damage index.
[0045] In one implementation, it further includes: An encoding module is used to input the labial gland tissue region and each of the sub-regions into a shared encoder to obtain the image features of the labial gland tissue region and each of the sub-regions output by the shared encoder; The first decoding module is used to input the image features of each of the sub-regions into the first decoder to obtain the number of lymph nodes and the area of each of the sub-regions output by the first decoder. The second decoding module is used to input the image features of the labial gland tissue region into the second decoder to obtain the inflammatory cell density and inflammatory area of the labial gland tissue region output by the second decoder; The third decoding module is used to input the image features of the labial gland tissue region into the third decoder to obtain the glandular atrophy area and total glandular area of the labial gland tissue region output by the third decoder.
[0046] The specific functions and examples of each module and submodule of the apparatus in this disclosure can be found in the relevant descriptions of the corresponding steps in the above method embodiments, and will not be repeated here.
[0047] The acquisition, storage, and application of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.
[0048] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.
[0049] Figure 4 This is a structural block diagram of an electronic device according to an embodiment of the present disclosure. Figure 4 As shown, the electronic device includes a memory 410 and a processor 420. The memory 410 stores a computer program that can run on the processor 420. There can be one or more memories 410 and processors 420. The memory 410 can store one or more computer programs, which, when executed by the electronic device, cause the electronic device to perform the methods provided in the above-described method embodiments. The electronic device may also include a communication interface 430 for communicating with external devices and performing data exchange and transmission.
[0050] If the memory 410, processor 420, and communication interface 430 are implemented independently, they can be interconnected via a bus to communicate with each other. This bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.
[0051] Optionally, in a specific implementation, if the memory 410, processor 420 and communication interface 430 are integrated on a single chip, the memory 410, processor 420 and communication interface 430 can communicate with each other through an internal interface.
[0052] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. General-purpose processors can be microprocessors or any conventional processor. It is worth noting that the processor can be a processor supporting Advanced Reduced Instruction Set Machines (ARM) architecture.
[0053] Further, optionally, the aforementioned memory may include read-only memory and random access memory, and may also include non-volatile random access memory. The memory may be volatile or non-volatile, or may include both. Non-volatile memory may include read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory may include random access memory (RAM), which serves as an external cache. Many forms of RAM are available by way of example, but not limitation. Examples include Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data SDRAM (DDR SDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct RAMBUS RAM (DR RAM).
[0054] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this disclosure are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line, DSL) or wireless (e.g., infrared, Bluetooth, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer, or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., Solid State Disks (SSDs)). It is worth noting that the computer-readable storage media mentioned in this disclosure may be non-volatile storage media; in other words, they may be non-transient storage media.
[0055] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.
[0056] In the description of the embodiments of this disclosure, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.
[0057] In the description of the embodiments disclosed herein, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone.
[0058] In the description of embodiments of this disclosure, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of embodiments of this disclosure, unless otherwise stated, "a plurality of" means two or more.
[0059] The above description is merely an exemplary embodiment of this disclosure and is not intended to limit this disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the protection scope of this disclosure.
Claims
1. A method for evaluating pathological images of labial glands in Sjögren's syndrome, characterized in that, include: The labial gland tissue region in the pathological image of the labial gland of the subject to be examined is divided into multiple sub-regions; Based on the pathological microenvironment characteristics of each sub-region, the weight coefficients of each sub-region are determined respectively; Based on the weight coefficients of each sub-region, the ratio of the number of lymphoid foci to the area of each sub-region is weighted and summed to obtain the lymphocyte foci score of the subject's dry mouth syndrome. Based on the distribution of lymph nodes in the labial gland tissue region across the various image sub-regions, a heterogeneity correction coefficient is determined. Based on the heterogeneity correction coefficient, the lymphocyte foci score of the subject's dry mouth syndrome is corrected.
2. The method according to claim 1, characterized in that, The determination of weight coefficients for each sub-region based on its respective pathological microenvironment characteristics includes: Based on the lymphocyte density, germinal center structure, and distance to adjacent ducts of the sub-region, a weight coefficient for the sub-region is determined; wherein, the lymphocyte density of the sub-region is positively correlated with the weight coefficient of the sub-region, whether the sub-region includes the germinal center structure is positively or negatively correlated with the value of the weight coefficient of the sub-region, and the distance to adjacent ducts of the sub-region is negatively correlated with the weight coefficient of the sub-region.
3. The method according to claim 2, characterized in that, The process of correcting the lymphocyte foci score for dry mouth syndrome in the subject of the examination based on the heterogeneity correction coefficient includes: The corrected lymphocyte score for dry mouth syndrome of the subject is determined by multiplying the heterogeneity correction coefficient with the lymphocyte score of dry mouth syndrome of the subject.
4. The method according to any one of claims 1-3, characterized in that, The labial gland tissue region in the pathological image of the subject's labial gland is divided into multiple sub-regions, including: The blank background and the labial gland tissue region in the labial gland pathology image of the subject to be examined are separated to obtain the labial gland tissue region in the labial gland pathology image; The labial gland tissue region is cropped using a sliding window to obtain multiple sub-regions; A color transfer algorithm is used to convert all the sub-regions to the same color space, resulting in the sub-regions after eliminating color differences.
5. The method according to claim 4, characterized in that, Also includes: Based on the inflammatory cell density and inflammatory area of the labial gland tissue region, the acute inflammation index of the dry labial gland syndrome of the subject to be tested is determined; Based on the proportion of glandular atrophy area in the labial gland tissue region to the total glandular area, and the degree of hardening of the ducts in the labial gland tissue region, the chronic damage index of dry labial gland syndrome in the subject to be tested is determined. Based on the modified lymphocyte foci score, acute inflammation index, and chronic damage index of the subject's dry mouth syndrome, the comprehensive assessment result of the subject's dry mouth syndrome was determined.
6. The method according to claim 5, characterized in that, Also includes: The labial gland tissue region and each of the sub-regions are respectively input into a shared encoder to obtain the image features of the labial gland tissue region and each of the sub-regions output by the shared encoder; The image features of each sub-region are input into the first decoder to obtain the number of lymph nodes and the area of each sub-region output by the first decoder. The image features of the labial gland tissue region are input into the second decoder to obtain the inflammatory cell density and inflammatory area of the labial gland tissue region output by the second decoder; The image features of the labial gland tissue region are input into the third decoder to obtain the glandular atrophy area and total glandular area of the labial gland tissue region output by the third decoder.
7. An evaluation device for pathological images of labial glands in Sjögren's syndrome, characterized in that, include: The region segmentation module is used to divide the labial gland tissue region in the labial gland pathology image of the subject to be examined into multiple sub-regions; The weight coefficient determination module is used to determine the weight coefficient of each sub-region based on the pathological microenvironment characteristics of each sub-region. The scoring determination module is used to perform a weighted summation of the ratio of the number of lymphoid foci to the area of each sub-region based on the weight coefficient of each sub-region, so as to obtain the lymphoid foci score of the subject's dry mouth syndrome. The correction coefficient determination module is used to determine the heterogeneity correction coefficient based on the distribution of lymph nodes in the labial gland tissue region in each of the image sub-regions; The scoring correction module is used to correct the lymphocyte foci score of the subject's dry mouth syndrome based on the heterogeneity correction coefficient.
8. The apparatus according to claim 7, characterized in that, The weight coefficient determination module is specifically used for: Based on the lymphocyte density, germinal center structure, and distance to adjacent ducts of the sub-region, a weight coefficient for the sub-region is determined; wherein, the lymphocyte density of the sub-region is positively correlated with the weight coefficient of the sub-region, whether the sub-region includes the germinal center structure is positively or negatively correlated with the value of the weight coefficient of the sub-region, and the distance to adjacent ducts of the sub-region is negatively correlated with the weight coefficient of the sub-region.
9. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.
10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-6.