Method and system for evaluating the effect of B cells on the prognosis of pancreatic cancer patients

Abstract

The present invention relates to a method and system for evaluating the effect of B cells on the prognosis of pancreatic cancer patients, wherein the method performs whole tissue section processing and CD20 immunohistochemical staining processing on historical PDAC specimens, scans the processed historical PDAC specimens by whole tissue section scanning technology, obtains whole tissue section images of historical PDAC specimens, and uses this to identify and extract B cells, generates a two-dimensional projection map based on all the obtained B cell coordinates, divides the two-dimensional projection map into an invasion front area, a cancer center area, and a cancer island area by a spatial analysis method, calculates the occupancy rate, fractal dimension, clustering density, and coverage rate of B cells in different areas, and obtains occupancy index, fractal index, clustering index, and coverage index. All indicators are combined with the survival cycle of pancreatic cancer patients for survival analysis, and the results of the effect of B cells on the prognosis of pancreatic cancer patients are obtained. Thus, the present invention improves the accuracy of evaluating the effect of B cells on the prognosis of pancreatic cancer patients.

Description

Technical Field

[0001] The present invention relates to the field of biological cell technology, and in particular to a method and system for evaluating the effect of B cells on the prognosis of pancreatic cancer patients. Background Art

[0002] In recent years, the impact of B cells on the prognosis of patients with different tumors has gradually begun to be explored. In the study of pancreatic cancer patients, the function of B cells in the tertiary lymphoid structure was analyzed mainly by using an optical microscope, which was significantly different from its function in the stromal microenvironment, and the research conclusion was drawn. However, the research results are contradictory. On the one hand, it is believed that the increase in B cell levels and the increase in their aggregates are related to improved prognosis. On the other hand, it is believed that the significant presence of B cells may lead to a worse prognosis by promoting the growth of pancreatic cancer. Therefore, the existing technology lacks an accurate evaluation method for the impact of B cells on the prognosis of pancreatic cancer patients. Summary of the invention

[0003] The technical problem to be solved by the present invention is: the present invention provides a method and system for evaluating the impact of B cells on the prognosis of pancreatic cancer patients, thereby improving the accuracy of the evaluation.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0005] In a first aspect, the present invention provides a method for evaluating the effect of B cells on the prognosis of pancreatic cancer patients, comprising:

[0006] Acquire a preset number of historical PDAC specimens removed by surgery, pre-process the historical PDAC specimens to obtain pre-processed historical PDAC specimens, wherein the pre-processing includes whole tissue section processing and CD20 immunohistochemical staining processing, scan the pre-processed historical PDAC specimens by whole tissue section scanning technology to obtain whole tissue section images of the historical PDAC specimens, identify and extract B cells from the whole tissue section images of the historical PDAC specimens by image segmentation method, and obtain each B cell and each B cell coordinate;

[0007] Generate a two-dimensional projection map based on all B cell coordinates, divide the two-dimensional projection map into an invasion front area, a cancer center area, and a cancer island area by a spatial analysis method, calculate the occupancy rate and fractal dimension of B cells in the invasion front area, the cancer center area, and the cancer island area, respectively, and obtain corresponding occupancy index and fractal index, and calculate the clustering density and coverage rate of B cells in the invasion front area, the cancer center area, and the cancer island area by a cluster analysis method, and obtain corresponding clustering index and coverage index;

[0008] The survival period of pancreatic cancer patients corresponding to the whole tissue section image of the historical PDAC specimen is obtained, and the survival period of the pancreatic cancer patients is combined with the occupancy index, the fractal index, the clustering index and the coverage index by using the Kaplan–Meier method to perform survival analysis to obtain a survival analysis result, and the survival analysis result is used as the result of the effect of B cells on the prognosis of pancreatic cancer patients.

[0009] The beneficial effects of the present invention are as follows: the whole tissue slice image of the historical PDAC specimen is obtained by scanning the historical PDAC specimen obtained after whole tissue slice processing and CD20 immunohistochemical staining by whole tissue slice scanning technology, and the whole tissue slice processing improves the integrity and comprehensiveness of the whole tissue slice image of the historical PDAC specimen obtained; the CD20 immunohistochemical staining improves the accuracy of B cell identification and extraction; the image segmentation method is used to realize the identification and extraction of B cells, and the automated process reduces manual operation and improves the evaluation efficiency; a two-dimensional projection map is generated based on all B cell coordinates, and it is divided into an invasion front area, a cancer center area and a cancer island area, and the fine-grained area division improves the comprehensiveness and integrity of the evaluation; the occupancy rate, fractal dimension, coverage density and coverage rate of B cells in different areas are calculated to obtain the occupancy index, fractal index, clustering index and coverage index, that is, multiple indicators are used to capture the spatial characteristics of B cells in different areas, and the spatial characteristics are combined with the survival cycle of pancreatic cancer patients for survival analysis, so as to improve the accuracy of the survival analysis results, that is, to improve the accuracy of the evaluation.

[0010] Optionally, obtaining a preset number of historical PDAC specimens removed by surgery includes:

[0011] Obtain case data corresponding to the historical PDAC specimens, and determine whether there are historical PDAC specimens that have received neoadjuvant therapy or radiotherapy in the historical PDAC specimens according to the case data; if so, filter the historical PDAC specimens that have received neoadjuvant therapy or radiotherapy to obtain filtered historical PDAC specimens.

[0012] According to the above description, the filtered historical PDAC specimens do not include cases that have received neoadjuvant therapy or radiotherapy, thus ensuring the integrity of the historical PDAC specimens.

[0013] Optionally, obtaining the filtered historical PDAC specimens comprises:

[0014] The PDAC resection surgery date and the end of life date are obtained from the case data corresponding to the filtered historical PDAC specimens, the date difference between the end of life date and the PDAC resection surgery date is calculated, and it is determined whether the date difference is lower than a date threshold. If so, the historical PDAC specimens lower than the date threshold are deleted to obtain the deleted historical PDAC specimens.

[0015] According to the above description, since the subsequent process of evaluating the impact of B cells on the prognosis of pancreatic cancer patients involves the survival cycle of pancreatic cancer patients, historical PDAC specimens whose date difference calculated based on the PDAC resection surgery date and the end of life date is lower than the date threshold will be deleted to avoid the impact of extreme situations on the evaluation and ensure the accuracy of the evaluation.

[0016] Optionally, the B cell identification and extraction of the whole tissue slice image of the historical PDAC specimen by image segmentation method to obtain each B cell and each B cell coordinate includes:

[0017] Performing downsampling processing on the historical PDAC specimen whole tissue slice image at a preset ratio to obtain the historical PDAC specimen whole tissue slice image after downsampling processing, and performing image cutting on the historical PDAC specimen whole tissue slice image after downsampling processing at a preset resolution to obtain image blocks with a preset resolution;

[0018] An unsupervised watershed algorithm is used to identify B cells in image blocks of preset resolution to obtain each B cell, and the coordinates of each B cell are extracted using QuPath technology to obtain the corresponding coordinates of each B cell.

[0019] According to the above description, the whole tissue section images of historical PDAC specimens were downsampled and cut with preset resolution, which reduced the storage space occupancy rate. The unsupervised watershed algorithm and QuPath technology were used for cell recognition and coordinate extraction to improve the accuracy of cell recognition and coordinate extraction.

[0020] Optionally, generating a two-dimensional projection map based on all B cell coordinates, and dividing the two-dimensional projection map into an invasion front area, a cancer center area, and a cancer island area by a spatial analysis method comprises:

[0021] A local coordinate system is formed based on all B cell coordinates, the local coordinate system is converted into a global coordinate system, a two-dimensional projection map is generated based on the global coordinate system, and in the process of generating the two-dimensional projection map, the boundaries of the overall tissue contour, the cancer tissue contour, and the normal tissue contour are annotated by ASAP software to obtain a labeled two-dimensional projection map;

[0022] According to the annotated two-dimensional projection image, the boundaries of the cancer tissue contour and the normal tissue contour are respectively expanded inward and outward by a first preset length distance to obtain an extended area, and the extended area is used as an invasion front area, and the cancer tissue contour is used as a cancer center area. The invasion front area and the cancer center area constitute a cancer island area.

[0023] According to the above description, the two-dimensional projection image is generated based on the global coordinate system, which ensures the integrity of the two-dimensional projection image. When performing zone division, the boundaries of the overall tissue contour, the cancer tissue contour and the normal tissue contour are first marked through the ASAP software, and then the corresponding zones are divided according to the marked boundaries to reduce the zone division error rate. When performing zone division, the invasion front area is taken into account, and the boundaries of the cancer tissue contour and the normal tissue contour are respectively expanded inwardly and outwardly by a first preset length to obtain an extended area as the invasion front area, which ensures the objectivity and accuracy of the obtained invasion front area while improving the comprehensiveness and accuracy of the cancer island area.

[0024] Optionally, the occupancy rate and fractal dimension of B cells in the invasion front area, the cancer center area and the cancer island area are calculated respectively to obtain corresponding occupancy index and fractal index, and the clustering density and coverage rate of B cells in the invasion front area, the cancer center area and the cancer island area are calculated by cluster analysis to obtain corresponding clustering index and coverage index, which include:

[0025] performing minimum bounding rectangle generation processing on the invasion front region, the cancer center region and the cancer island region respectively to obtain corresponding invasion front bounding rectangle, cancer center bounding rectangle and cancer island bounding rectangle;

[0026] Performing square grid division of preset scale in the invasion front circumscribed rectangle, the cancer center circumscribed rectangle and the cancer island circumscribed rectangle respectively to obtain the invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division and the cancer island circumscribed rectangle after square grid division, wherein the preset scale includes a first scale and a second scale, and the first scale is different from the second scale;

[0027] The occupancy rates and analysis dimensions of the B cell invasion front bounding rectangle after square grid division, the cancer center bounding rectangle after square grid division, and the cancer island bounding rectangle after square grid division were calculated respectively. At the same time, the clustering density and coverage rate of the B cell invasion front bounding rectangle after square grid division, the cancer center bounding rectangle after square grid division, and the cancer island bounding rectangle after square grid division were calculated by cluster analysis method to obtain the corresponding clustering index and coverage index.

[0028] Optionally, the occupancy rate and analysis dimension of the circumscribed rectangle of the invasion front of B cells after square grid division, the circumscribed rectangle of the cancer center after square grid division, and the circumscribed rectangle of the cancer island after square grid division are calculated respectively, and the clustering density and coverage rate of the circumscribed rectangle of the invasion front of B cells after square grid division, the circumscribed rectangle of the cancer center after square grid division, and the circumscribed rectangle of the cancer island after square grid division are calculated by cluster analysis method, and the corresponding clustering index and coverage index are obtained, including:

[0029] The occupancy rate of B cells in the invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division were calculated by the occupancy rate formula to obtain the corresponding occupancy index, where the occupancy rate formula is:

[0030]

[0031] Among them, OR represents the occupancy rate, represents the number of square grids occupied by B cells, represents the number of square grids occupied by non-B cells;

[0032] The fractal dimension formula was used to calculate the fractal dimensions of the B cells in the circumscribed rectangle of the invasion front after square grid division, the circumscribed rectangle of the cancer center after square grid division, and the circumscribed rectangle of the cancer island after square grid division, and the fractal index was obtained. The fractal dimension formula is:

[0033]

[0034] Where FD represents the fractal dimension, S represents the slope, represents the area of ​​all square grids, d=2, represents a two-dimensional plane, and A represents a constant;

[0035] HDBSCAN clustering algorithm is used to cluster B cells in the invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division according to the preset minimum clustering scale and the preset clustering strictness, and the corresponding clusters are obtained;

[0036] The clustering density formula is used to calculate the clustering density of clusters in the corresponding invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division to obtain the clustering index, where the clustering density formula is:

[0037]

[0038] in, The number of clusters, Indicates the area where the cluster is located;

[0039] The coverage rate of the clusters in the corresponding square grid-divided invasion front circumscribed rectangle, square grid-divided cancer center circumscribed rectangle, and square grid-divided cancer island circumscribed rectangle is calculated by the coverage rate formula to obtain the coverage index, where the coverage rate formula is:

[0040]

[0041] in, represents the coverage rate, represents the area of ​​the cluster, Indicates the area of ​​the region where the cluster is located.

[0042] According to the above description, when calculating the occupancy rate, fractal dimension, clustering density and coverage rate, the minimum bounding rectangle generation process is first performed on each area to ensure that when the square grid division of the preset scale is performed within the corresponding bounding rectangle, the square grid can completely cover the corresponding area, that is, the invasion front area, the cancer center area and the cancer island area are completely covered by the corresponding square grid, which ensures the accuracy of calculating the occupancy rate, fractal dimension, clustering density and coverage rate of B cells in combination with the grid side length, area and number of the square grid, and when the square grid division is performed, the square grid division of different preset scales is performed, that is, the spatial characteristics of B cells at different preset scales are fully considered to improve the integrity of the obtained occupancy index, fractal index, clustering index and coverage index.

[0043] Optionally, the survival analysis is performed by combining the pancreatic cancer patient's survival period with the occupancy index, the fractal index, the clustering index, and the coverage index using the Kaplan–Meier method, and the survival analysis result obtained includes:

[0044] Determine whether the survival period of the pancreatic cancer patient is lower than a period threshold, if so, mark the survival period of the pancreatic cancer patient as short-term, otherwise, mark the survival period of the adenocarcinoma patient as long-term;

[0045] The survival cycles of all adenocarcinoma patients marked as short-term were summarized to obtain the short-term survival group, and the survival cycles of all adenocarcinoma patients marked as long-term were summarized to obtain the long-term survival group;

[0046] The Kaplan–Meier method was used to calculate the survival probability curves of the short-term survival group and the long-term survival group, and the log-rank test method was used to compare the median difference values ​​of the occupancy index, fractal index, clustering index and coverage index in the survival probability curve of the long-term survival group with those in the survival probability curve of the short-term survival group;

[0047] When the median difference value exceeds the difference threshold, a survival analysis result is obtained, indicating that the indicator corresponding to the median difference value exceeding the difference threshold has a positive impact on the prognosis of pancreatic cancer patients. Conversely, when the median difference value is lower than the difference threshold, a survival analysis result is obtained, indicating that the indicator corresponding to the median difference value lower than the difference threshold has a negative impact on the prognosis of pancreatic cancer patients.

[0048] Optionally, the cycle threshold is 24 months.

[0049] According to the above description, when the survival period of pancreatic cancer patients is combined with the occupancy index, fractal index, clustering index and coverage index for survival analysis, the survival period of pancreatic cancer patients is first grouped into long-term and short-term groups, and the survival probability curves of different groups are calculated respectively, that is, the survival probability curve of the short-term survival group and the survival probability curve of the long-term survival group. In this way, the log-rank test method is used to compare the median difference values ​​of the occupancy index, fractal index, clustering index and coverage index in different survival probability curves, so as to improve the accuracy and rationality of the survival analysis results obtained according to the median difference value.

[0050] In a second aspect, the present invention provides a system for evaluating the impact of B cells on the prognosis of pancreatic cancer patients, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for evaluating the impact of B cells on the prognosis of pancreatic cancer patients described in the first aspect is implemented.

[0051] Among them, the technical effect corresponding to the system for evaluating the impact of B cells on the prognosis of pancreatic cancer patients provided by the second aspect refers to the relevant description of the method for evaluating the impact of B cells on the prognosis of pancreatic cancer patients provided by the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 A flow chart of the method for evaluating the effect of B cells on the prognosis of pancreatic cancer patients provided in this embodiment;

[0053] Figure 2 This is a schematic diagram of the overall process of the method for evaluating the effect of B cells on the prognosis of pancreatic cancer patients provided in this embodiment;

[0054] Figure 3is a schematic diagram of the survival analysis process involved in this embodiment;

[0055] Figure 4 This is a schematic diagram of the structure of the system for evaluating the impact of B cells on the prognosis of pancreatic cancer patients provided in this embodiment.

[0056] Description of reference numerals:

[0057] 1. A system to evaluate the impact of B cells on the prognosis of pancreatic cancer patients;

[0058] 2. Processor;

[0059] 3. Memory. DETAILED DESCRIPTION

[0060] In order to better understand the above technical solution, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0061] Embodiment 1

[0062] Please refer to Figures 1 to 3 The present invention provides a method for evaluating the effect of B cells on the prognosis of pancreatic cancer patients, comprising the steps of:

[0063] S1. Obtain a preset number of historical PDAC specimens removed by surgery, pre-process the historical PDAC specimens to obtain pre-processed historical PDAC specimens, wherein the pre-processing includes whole tissue section processing and CD20 immunohistochemical staining processing, scan the pre-processed historical PDAC specimens by whole tissue section scanning technology to obtain whole tissue section images of the historical PDAC specimens, identify and extract B cells from the whole tissue section images of the historical PDAC specimens by image segmentation method, and obtain each B cell and each B cell coordinate;

[0064] In this embodiment, if Figure 2 As shown, a preset number of historical PDAC (Pancreatic Ductal Adenocarcinoma) specimens and their corresponding case data are obtained after surgical resection. The preset number is 40-80, and preferably, the preset number is 45. The case data includes but is not limited to: age of pancreatic cancer patients, gender of pancreatic cancer patients, date of PDAC resection surgery, date of death, whether neoadjuvant therapy has been received, whether radiotherapy has been received, PN stage and PT stage.

[0065] Among them, the pretreatment includes whole tissue section processing and CD20 immunohistochemical staining processing. The historical PDAC specimens are fixed with formalin and embedded in paraffin. When the whole tissue section is processed, the whole tissue section is processed according to the preset thickness, which is 4 microns, that is, the historical PDAC specimens are processed with a 4-micron thickness to obtain all the historical PDAC sections, and each historical PDAC section is stained with H&E to obtain the historical PDAC sections after H&E staining; the representative historical PDAC sections containing tumor tissue and its surrounding healthy tissue are screened from the historical PDAC sections after H&E staining, and the screened historical PDAC sections are stained with CD20 immunohistochemical staining to obtain the historical PDAC sections after CD20 immunohistochemical staining; the historical PDAC sections after CD20 immunohistochemical staining are scanned by the whole tissue section scanning technology to obtain the whole tissue section image of the historical PDAC specimen, and the whole tissue section image of the historical PDAC specimen is identified and extracted by image segmentation. Among them, B cells specifically refer to: CD20+ B cells, and each B cell and each B cell coordinate are obtained.

[0066] At this time, obtaining a preset number of historical PDAC specimens removed by surgery in step S1 includes:

[0067] S11, obtaining case data corresponding to the historical PDAC specimens, and determining whether there are historical PDAC specimens that have received neoadjuvant therapy or radiotherapy in the historical PDAC specimens according to the case data; if so, filtering the historical PDAC specimens that have received neoadjuvant therapy or radiotherapy to obtain filtered historical PDAC specimens.

[0068] In this embodiment, if Figure 2 As shown, historical PDAC specimens that had received neoadjuvant therapy or radiotherapy were filtered to obtain filtered historical PDAC specimens.

[0069] At this time, the filtered historical PDAC specimens obtained in step S11 include:

[0070] S111, obtaining the PDAC resection surgery date and the end of life date from the case data corresponding to the filtered historical PDAC specimens, calculating the date difference between the end of life date and the PDAC resection surgery date, and determining whether the date difference is lower than a date threshold, and if so, deleting the historical PDAC specimens lower than the date threshold to obtain deleted historical PDAC specimens.

[0071] In this embodiment, if Figure 2As shown, before preprocessing the filtered historical PDAC specimens, the date difference between the end of life date and the PDAC resection surgery date is calculated. If the date difference is lower than the date threshold, where the date threshold is 31 days, that is, if the date difference is lower than 31 days, the corresponding historical PDAC specimen is deleted to obtain the deleted historical PDAC specimen, and the deleted historical PDAC specimen is preprocessed. The date threshold can be adjusted according to actual conditions.

[0072] At this time, the image segmentation method is used in step S1 to identify and extract B cells from the whole tissue slice image of the historical PDAC specimen, and each B cell and each B cell coordinate are obtained, including:

[0073] S12, performing downsampling processing on the historical PDAC specimen whole tissue slice image at a preset ratio to obtain the historical PDAC specimen whole tissue slice image after downsampling processing, and performing image cutting on the historical PDAC specimen whole tissue slice image after downsampling processing at a preset resolution to obtain image blocks with a preset resolution;

[0074] In this embodiment, if Figure 2 As shown, the preset ratio is one-fourth, and the preset resolution is 400*400 pixels, that is, the whole tissue slice image of the historical PDAC specimen is downsampled by one-fourth to obtain the whole tissue slice image of the historical PDAC specimen after downsampling, and the whole tissue slice image of the historical PDAC specimen after downsampling is cut into 400*400 pixels, that is, the whole tissue slice image of the historical PDAC specimen after downsampling is cut into image blocks with a resolution of 400*400 pixels, and the image cutting can be implemented by using a Groovy script. The preset ratio and preset resolution can be adjusted according to actual conditions.

[0075] S13. Use an unsupervised watershed algorithm to identify B cells on image blocks of preset resolution to obtain each B cell, and use QuPath technology to extract the coordinates of each B cell to obtain the corresponding coordinates of each B cell.

[0076] In this embodiment, if Figure 2 As shown, an unsupervised watershed algorithm is used for B cell recognition, and during the recognition process, a high threshold is set to remove background noise and nonspecific immune staining to obtain each B cell, and the coordinates of each B cell are extracted using QuPath technology to obtain the coordinates of each B cell.

[0077] S2. Generate a two-dimensional projection map based on all B cell coordinates, divide the two-dimensional projection map into an invasion front area, a cancer center area, and a cancer island area by a spatial analysis method, calculate the occupancy rate and fractal dimension of B cells in the invasion front area, the cancer center area, and the cancer island area, respectively, and obtain corresponding occupancy index and fractal index, and calculate the clustering density and coverage rate of B cells in the invasion front area, the cancer center area, and the cancer island area by a cluster analysis method, and obtain corresponding clustering index and coverage index;

[0078] At this time, the two-dimensional projection map generated based on all B cell coordinates in step S2, and the two-dimensional projection map is divided into the invasion front area, the cancer center area and the cancer island area by the spatial analysis method, including:

[0079] S21, constructing a local coordinate system based on all B cell coordinates, converting the local coordinate system into a global coordinate system, generating a two-dimensional projection map based on the global coordinate system, and in the process of generating the two-dimensional projection map, annotating the boundaries of the overall tissue contour, the cancer tissue contour, and the normal tissue contour by using ASAP software to obtain the annotated two-dimensional projection map;

[0080] In this embodiment, if Figure 2 As shown, since the B cell coordinates are obtained based on image blocks with preset resolution, the coordinate system constructed by all B cell coordinates is a local coordinate system. The local coordinate system is converted into a global coordinate system through an existing coordinate conversion algorithm, so as to generate a two-dimensional projection map according to the global coordinate system, that is, to construct a B cell map. In the process of generating the two-dimensional projection map, the boundaries of the overall tissue contour, cancer tissue contour and normal tissue contour are annotated by ASAP software to obtain the annotated two-dimensional projection map.

[0081] S22. According to the annotated two-dimensional projection image, the boundaries of the cancer tissue contour and the normal tissue contour are respectively expanded inward and outward by a first preset length distance to obtain an extended area, and the extended area is used as an invasion front area, and the cancer tissue contour is used as a cancer center area. The invasion front area and the cancer center area constitute a cancer island area.

[0082] In this embodiment, if Figure 2 As shown, since the tissue contour is a closed shape, the cancer tissue contour is directly used as the cancer center area, and the boundaries of the cancer tissue contour and the normal tissue contour are respectively extended inward and outward by a first preset length distance to obtain an extended area, and the extended area is used as the invasion front area, wherein the first preset length distance is 0.5 mm, and the cancer island area includes the invasion front area and the cancer center area, that is, the invasion front area and the cancer center area constitute the cancer island area, wherein the first preset length distance can be adjusted according to actual conditions.

[0083] At this time, the occupancy rate and fractal dimension of B cells in the invasion front area, the cancer center area and the cancer island area are calculated respectively in step S2 to obtain corresponding occupancy index and fractal index, and the clustering density and coverage rate of B cells in the invasion front area, the cancer center area and the cancer island area are calculated by cluster analysis to obtain corresponding clustering index and coverage index, including:

[0084] S23, performing minimum bounding rectangle generation processing on the invasion front area, the cancer center area and the cancer island area respectively to obtain corresponding invasion front bounding rectangle, cancer center bounding rectangle and cancer island bounding rectangle;

[0085] In this embodiment, if Figure 2 As shown, with the help of Python's shapely library, the minimum bounding rectangles of the invasion front area, cancer center area and cancer island area are generated respectively to obtain the corresponding invasion front bounding rectangle, cancer center bounding rectangle and cancer island bounding rectangle.

[0086] S24, performing square grid division of a preset scale in the invasion front circumscribed rectangle, the cancer center circumscribed rectangle, and the cancer island circumscribed rectangle, respectively, to obtain the invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division, wherein the preset scale includes a first scale and a second scale, and the first scale is different from the second scale;

[0087] In this embodiment, if Figure 2 As shown, square grid divisions of the first scale and the second scale are performed in the invasion front circumscribed rectangle, the cancer center circumscribed rectangle and the cancer island circumscribed rectangle respectively to obtain the invasion front circumscribed rectangle after square grid divisions of different scales, the cancer center circumscribed rectangle after square grid divisions of different scales and the cancer island circumscribed rectangle after square grid divisions of different scales, wherein the first scale is 10-40 microns and the second scale is 200-600 microns. Preferably, the first scale is 40 microns and the second scale is 600 microns. Specifically, the first scale and the second scale can be adjusted according to actual conditions.

[0088] S25. Calculate the occupancy rate and analysis dimension of the circumscribed rectangle of the invasion front of B cells after square grid division, the circumscribed rectangle of the cancer center after square grid division, and the circumscribed rectangle of the cancer island after square grid division respectively. At the same time, calculate the clustering density and coverage rate of the circumscribed rectangle of the invasion front of B cells after square grid division, the circumscribed rectangle of the cancer center after square grid division, and the circumscribed rectangle of the cancer island after square grid division by cluster analysis method to obtain the corresponding clustering index and coverage index.

[0089] At this time, step S25 includes:

[0090] S251. Calculate the occupancy rate of B cells in the invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division respectively by using the occupancy rate formula to obtain the corresponding occupancy index, wherein the occupancy rate formula is:

[0091]

[0092] Among them, OR represents the occupancy rate, represents the number of square grids occupied by B cells, represents the number of square grids occupied by non-B cells;

[0093] In this embodiment, if Figure 2 As shown, the occupancy rates of B cells in the circumscribed rectangle of the invasion front after square grid division, the circumscribed rectangle of the cancer center after square grid division, and the circumscribed rectangle of the cancer island after square grid division are calculated respectively by the occupancy formula, that is, the occupancy rate is obtained by calculating the ratio of the number of square grids occupied by B cells to the total number of square grids, and the obtained occupancy rate is used as the occupancy index.

[0094] S252. Calculate the fractal dimensions of the invasion front circumscribed rectangle of B cells after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division by using the fractal dimension formula to obtain a fractal index, wherein the fractal dimension formula is:

[0095]

[0096] Where FD represents the fractal dimension, S represents the slope, represents the area of ​​all square grids, d=2, represents a two-dimensional plane, and A represents a constant;

[0097] In this embodiment, if Figure 2 As shown, the fractal dimensions of B cells in the circumscribed rectangle of the invasion front after square grid division, the circumscribed rectangle of the cancer center after square grid division, and the circumscribed rectangle of the cancer island after square grid division are calculated respectively by the fractal dimension formula, that is, the fractal dimension is obtained by fitting the set of B cells by least square linear regression and the slope of the fitting line, and the obtained fractal dimension is used as the fractal index. In the fractal dimension formula, when the tissue contours at different scales are similar, the B cell distribution follows the power law formula:

[0098] = .

[0099] S253, using the HDBSCAN clustering algorithm to cluster the B cells in the invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division according to a preset minimum clustering scale and a preset clustering strictness, to obtain corresponding clusters;

[0100] In this embodiment, if Figure 2 As shown, the preset minimum clustering scale is 30, the preset clustering strictness is 15, and the HDBSCAN clustering algorithm is used to cluster the B cells in the invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division according to the minimum clustering scale of 30 and the clustering strictness of 15 to obtain the corresponding cluster clusters.

[0101] S254, calculating the clustering density of the clusters in the corresponding invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division by using the clustering density formula to obtain a clustering index, wherein the clustering density formula is:

[0102]

[0103] in, The number of clusters, Indicates the area where the cluster is located;

[0104] S255. Calculate the coverage of the clusters in the corresponding invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division by using the coverage formula to obtain a coverage index, wherein the coverage formula is:

[0105]

[0106] in, represents the coverage rate, represents the area of ​​the cluster, Indicates the area where the cluster is located.

[0107] In this embodiment, if Figure 2 As shown, the clustering density formula and the coverage formula are used to calculate the clustering density and coverage of the clusters in the corresponding invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division, respectively, so as to obtain the corresponding clustering index and coverage index.

[0108] S3. Obtain the pancreatic cancer patient survival period corresponding to the whole tissue section image of the historical PDAC specimen, and use the Kaplan–Meier method to combine the pancreatic cancer patient survival period with the occupancy index, the fractal index, the clustering index, and the coverage index to perform survival analysis to obtain a survival analysis result, and use the survival analysis result as a result of the effect of B cells on the prognosis of pancreatic cancer patients.

[0109] In this embodiment, if Figure 2 As shown, the survival period of pancreatic cancer patients corresponding to the whole tissue section images of historical PDAC specimens is obtained, wherein the survival period of pancreatic cancer patients refers to the time period from the diagnosis date of the first diagnosis of PDAC to the death date of death due to any cause. The Kaplan–Meier method is used to combine the survival period of pancreatic cancer patients with the occupancy index, fractal index, clustering index and coverage index to perform survival analysis, thereby obtaining the survival analysis result, and the survival analysis result is used as the result of the influence of B cells on the prognosis of pancreatic cancer patients.

[0110] At this time, the survival analysis is performed by combining the pancreatic cancer patient's survival period with the occupancy index, the fractal index, the clustering index, and the coverage index using the Kaplan–Meier method in step S3, and the survival analysis results obtained include:

[0111] S31, determining whether the survival period of the pancreatic cancer patient is lower than a period threshold, if so, marking the survival period of the pancreatic cancer patient as short-term, otherwise, marking the survival period of the adenocarcinoma patient as long-term;

[0112] S32, summing up the survival cycles of all adenocarcinoma patients marked as short-term to obtain a short-term survival group, and summing up the survival cycles of all adenocarcinoma patients marked as long-term to obtain a long-term survival group;

[0113] In this embodiment, if Figure 3 As shown, the cycle threshold is 24 months. The survival cycle of pancreatic cancer patients whose survival cycle is lower than the cycle threshold is marked as short-term, and the survival cycle of pancreatic cancer patients whose survival cycle is not lower than the cycle threshold is marked as long-term. Thus, all the survival cycles of adenocarcinoma patients marked as short-term are summarized to obtain a short-term survival group, and all the survival cycles of adenocarcinoma patients marked as long-term are summarized to obtain a long-term survival group.

[0114] S33, using the Kaplan–Meier method to calculate the survival probability curves of the short-term survival group and the long-term survival group, respectively, and using the log-rank test method to compare the median difference values ​​between the occupancy index, fractal index, clustering index and coverage index in the survival probability curve of the long-term survival group and the occupancy index, fractal index, clustering index and coverage index in the survival probability curve of the short-term survival group;

[0115] S34. When the median difference value exceeds the difference threshold, the survival analysis result is obtained, indicating that the indicator corresponding to the median difference value exceeding the difference threshold has a positive impact on the prognosis of pancreatic cancer patients. Conversely, when the median difference value is lower than the difference threshold, the survival analysis result is obtained, indicating that the indicator corresponding to the median difference value lower than the difference threshold has a negative impact on the prognosis of pancreatic cancer patients.

[0116] In this embodiment, if Figure 3 As shown, the Kaplan–Meier method was used to calculate the survival probability curves of the short-term survival group and the long-term survival group, and the log-rank test method was used to compare the median differences of the occupancy index, fractal index, clustering index and coverage index in the survival probability curve of the long-term survival group with those in the survival probability curve of the short-term survival group, that is, the median difference of the occupancy index was obtained by subtracting the median of the occupancy index in the short-term survival probability curve from the median of the occupancy index in the survival probability curve of the long-term survival group, the median difference of the fractal index was obtained by subtracting the median of the fractal index in the short-term survival probability curve from the median of the fractal index in the survival probability curve of the long-term survival group, the median difference of the clustering index was obtained by subtracting the median of the clustering index in the short-term survival probability curve from the median of the clustering index in the survival probability curve of the long-term survival group, and the median difference of the coverage index was obtained by subtracting the median of the coverage index in the short-term survival probability curve from the median of the coverage index in the survival probability curve of the long-term survival group. If the median difference value exceeds the difference threshold, where the difference threshold is 0.05, that is, if the median difference value exceeds 0.05, it means that the indicator corresponding to the median difference value exceeding the difference threshold has a positive effect on the prognosis of pancreatic cancer patients, that is, the survival analysis result that the indicator has a positive effect on the prognosis of pancreatic cancer patients is obtained. Conversely, if the median difference value is lower than 0.05, it means that the indicator corresponding to the median difference value lower than the difference threshold has a negative effect on the prognosis of pancreatic cancer patients, that is, the survival analysis result that the indicator has a negative effect on the prognosis of pancreatic cancer patients is obtained.

[0117] Embodiment 2

[0118] Please refer to Figure 4 The present invention provides a system 1 for evaluating the impact of B cells on the prognosis of pancreatic cancer patients, comprising a memory 3, a processor 2, and a computer program stored in the memory 3 and executable on the processor 2. When the processor 2 executes the computer program, the steps in Example 1 are implemented.

[0119] Since the system / device described in the above embodiments of the present invention is a system / device used to implement the method of the above embodiments of the present invention, a person skilled in the art can understand the specific structure and deformation of the system / device based on the method described in the above embodiments of the present invention, and thus will not be described in detail here. All systems / devices used in the method of the above embodiments of the present invention belong to the scope of protection of the present invention.

[0120] It should be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0121] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions.

[0122] It should be noted that in the claims, any reference numerals placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In the claims enumerating several means, several of these means may be embodied by the same hardware. The use of the words first, second, third, etc., is for convenience of expression only and does not indicate any order. These words may be understood as part of the component name.

[0123] In addition, it should be noted that, in the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0124] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments after knowing the basic creative concept. Therefore, the claims should be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0125] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include these modifications and variations.

Claims

1. A method for evaluating the effect of B cells on the prognosis of pancreatic cancer patients, characterized in that: include: Obtaining a preset number of historical pancreatic ductal adenocarcinoma (PDAC) specimens that were surgically removed, pre-processing the historical pancreatic ductal adenocarcinoma (PDAC) specimens to obtain pre-processed historical pancreatic ductal adenocarcinoma (PDAC) specimens, wherein the pre-processing includes whole tissue section processing and CD20 immunohistochemical staining processing, scanning the pre-processed historical pancreatic ductal adenocarcinoma (PDAC) specimens by whole tissue section scanning technology to obtain whole tissue section images of the historical pancreatic ductal adenocarcinoma (PDAC) specimens, and performing B cell identification and extraction on the whole tissue section images of the historical pancreatic ductal adenocarcinoma (PDAC) specimens by an image segmentation method to obtain each B cell and each B cell coordinate; Generate a two-dimensional projection map based on all B cell coordinates, divide the two-dimensional projection map into an invasion front area, a cancer center area, and a cancer island area by a spatial analysis method, calculate the occupancy rate and fractal dimension of B cells in the invasion front area, the cancer center area, and the cancer island area, respectively, and obtain corresponding occupancy index and fractal index, and calculate the clustering density and coverage rate of B cells in the invasion front area, the cancer center area, and the cancer island area by a cluster analysis method, and obtain corresponding clustering index and coverage index; Obtaining the pancreatic cancer patient survival period corresponding to the historical pancreatic ductal adenocarcinoma PDAC specimen whole tissue slice image, combining the pancreatic cancer patient survival period with the occupancy index, the fractal index, the clustering index, and the coverage index using the Kaplan–Meier method to perform survival analysis to obtain a survival analysis result, and using the survival analysis result as a result of the effect of B cells on the prognosis of pancreatic cancer patients; The survival analysis is performed by combining the pancreatic cancer patient's survival period with the occupancy index, the fractal index, the clustering index, and the coverage index using the Kaplan–Meier method, and the survival analysis results obtained include: Determine whether the survival period of the pancreatic cancer patient is lower than a period threshold, if so, mark the survival period of the pancreatic cancer patient as short-term, otherwise, mark the survival period of the adenocarcinoma patient as long-term; The survival cycles of all adenocarcinoma patients marked as short-term were summarized to obtain the short-term survival group, and the survival cycles of all adenocarcinoma patients marked as long-term were summarized to obtain the long-term survival group; The Kaplan–Meier method was used to calculate the survival probability curves of the short-term survival group and the long-term survival group, and the log-rank test method was used to compare the median difference values ​​of the occupancy index, fractal index, clustering index and coverage index in the survival probability curve of the long-term survival group with those in the survival probability curve of the short-term survival group; When the median difference value exceeds the difference threshold, a survival analysis result is obtained, indicating that the indicator corresponding to the median difference value exceeding the difference threshold has a positive impact on the prognosis of pancreatic cancer patients. Conversely, when the median difference value is lower than the difference threshold, a survival analysis result is obtained, indicating that the indicator corresponding to the median difference value lower than the difference threshold has a negative impact on the prognosis of pancreatic cancer patients.

2. The method for evaluating the effect of B cells on the prognosis of pancreatic cancer patients according to claim 1, characterized in that: The procedure involves obtaining a predetermined number of historical pancreatic ductal adenocarcinoma (PDAC) specimens from surgical resections, including: Obtain case data corresponding to the historical pancreatic ductal adenocarcinoma PDAC specimens, and determine, according to the case data, whether there are historical pancreatic ductal adenocarcinoma PDAC specimens that have received neoadjuvant therapy or radiotherapy in the historical pancreatic ductal adenocarcinoma PDAC specimens; if so, filter the historical pancreatic ductal adenocarcinoma PDAC specimens that have received neoadjuvant therapy or radiotherapy to obtain filtered historical pancreatic ductal adenocarcinoma PDAC specimens.

3. The method for evaluating the effect of B cells on the prognosis of pancreatic cancer patients according to claim 2, characterized in that: The filtered historical pancreatic ductal adenocarcinoma (PDAC) specimens include: The pancreatic ductal adenocarcinoma PDAC resection surgery date and the end of life date are obtained from the case data corresponding to the filtered historical pancreatic ductal adenocarcinoma PDAC specimens, the date difference between the end of life date and the pancreatic ductal adenocarcinoma PDAC resection surgery date is calculated, and it is determined whether the date difference is lower than a date threshold. If so, the historical pancreatic ductal adenocarcinoma PDAC specimens lower than the date threshold are deleted to obtain the deleted historical pancreatic ductal adenocarcinoma PDAC specimens.

4. The method for evaluating the effect of B cells on the prognosis of pancreatic cancer patients according to claim 1, characterized in that: The image segmentation method is used to identify and extract B cells from the whole tissue slice image of the historical pancreatic ductal adenocarcinoma PDAC specimen, and each B cell and each B cell coordinate are obtained, including: Performing downsampling processing on the historical pancreatic ductal adenocarcinoma (PDAC) specimen whole tissue slice image at a preset ratio to obtain the downsampling processed historical pancreatic ductal adenocarcinoma (PDAC) specimen whole tissue slice image, and performing image cutting on the downsampling processed historical pancreatic ductal adenocarcinoma (PDAC) specimen whole tissue slice image at a preset resolution to obtain image blocks with a preset resolution; An unsupervised watershed algorithm is used to identify B cells in image blocks of preset resolution to obtain each B cell, and the coordinates of each B cell are extracted using QuPath technology to obtain the corresponding coordinates of each B cell.

5. The method for evaluating the effect of B cells on the prognosis of pancreatic cancer patients according to claim 1, characterized in that: Generating a two-dimensional projection map based on all B cell coordinates, and dividing the two-dimensional projection map into an invasion front area, a cancer center area, and a cancer island area by a spatial analysis method comprises: A local coordinate system is formed based on all B cell coordinates, the local coordinate system is converted into a global coordinate system, a two-dimensional projection map is generated based on the global coordinate system, and in the process of generating the two-dimensional projection map, the boundaries of the overall tissue contour, the cancer tissue contour, and the normal tissue contour are annotated by ASAP software to obtain a labeled two-dimensional projection map; According to the annotated two-dimensional projection image, the boundaries of the cancer tissue contour and the normal tissue contour are respectively expanded inward and outward by a first preset length distance to obtain an extended area, and the extended area is used as an invasion front area, and the cancer tissue contour is used as a cancer center area. The invasion front area and the cancer center area constitute a cancer island area.

6. The method for evaluating the effect of B cells on the prognosis of pancreatic cancer patients according to claim 1, characterized in that: The occupancy rate and fractal dimension of B cells in the invasion front area, the cancer center area and the cancer island area are calculated respectively to obtain corresponding occupancy index and fractal index, and the clustering density and coverage rate of B cells in the invasion front area, the cancer center area and the cancer island area are calculated by cluster analysis to obtain corresponding clustering index and coverage index, which include: performing minimum bounding rectangle generation processing on the invasion front region, the cancer center region and the cancer island region respectively to obtain corresponding invasion front bounding rectangle, cancer center bounding rectangle and cancer island bounding rectangle; Performing square grid division of preset scale in the invasion front circumscribed rectangle, the cancer center circumscribed rectangle and the cancer island circumscribed rectangle respectively to obtain the invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division and the cancer island circumscribed rectangle after square grid division, wherein the preset scale includes a first scale and a second scale, and the first scale is different from the second scale; The occupancy rates and analysis dimensions of the B cell invasion front bounding rectangle after square grid division, the cancer center bounding rectangle after square grid division, and the cancer island bounding rectangle after square grid division were calculated respectively. At the same time, the clustering density and coverage rate of the B cell invasion front bounding rectangle after square grid division, the cancer center bounding rectangle after square grid division, and the cancer island bounding rectangle after square grid division were calculated by cluster analysis method to obtain the corresponding clustering index and coverage index.

7. The method for evaluating the effect of B cells on the prognosis of pancreatic cancer patients according to claim 1, characterized in that: The occupancy rate and analysis dimension of the circumscribed rectangle of the invasion front of B cells after the square grid division, the circumscribed rectangle of the cancer center after the square grid division, and the circumscribed rectangle of the cancer island after the square grid division are calculated respectively, and the clustering density and coverage rate of the circumscribed rectangle of the invasion front of B cells after the square grid division, the circumscribed rectangle of the cancer center after the square grid division, and the circumscribed rectangle of the cancer island after the square grid division are calculated by cluster analysis method, and the corresponding clustering index and coverage index are obtained, including: The occupancy rate of B cells in the invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division were calculated by the occupancy rate formula to obtain the corresponding occupancy index, where the occupancy rate formula is: Among them, OR represents the occupancy rate, represents the number of square grids occupied by B cells, represents the number of square grids occupied by non-B cells; The fractal dimension formula was used to calculate the fractal dimensions of the B cells in the circumscribed rectangle of the invasion front after square grid division, the circumscribed rectangle of the cancer center after square grid division, and the circumscribed rectangle of the cancer island after square grid division, and the fractal index was obtained. The fractal dimension formula is: Where FD represents the fractal dimension, S represents the slope, represents the area of ​​all square grids, d=2, represents a two-dimensional plane, and A represents a constant; HDBSCAN clustering algorithm is used to cluster B cells in the invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division according to the preset minimum clustering scale and the preset clustering strictness, and the corresponding clusters are obtained; The clustering density formula is used to calculate the clustering density of clusters in the corresponding invasion front circumscribed rectangle after square grid division, the cancer center circumscribed rectangle after square grid division, and the cancer island circumscribed rectangle after square grid division to obtain the clustering index, where the clustering density formula is: in, represents the cluster density, represents the number of clusters, Indicates the area where the cluster is located; The coverage rate formula is used to calculate the coverage rate of the clusters in the corresponding square grid-divided invasion front circumscribed rectangle, square grid-divided cancer center circumscribed rectangle, and square grid-divided cancer island circumscribed rectangle to obtain the coverage index, where the coverage rate formula is: in, represents the coverage rate, represents the area of ​​the cluster, Indicates the area where the cluster is located.

8. The method for evaluating the effect of B cells on the prognosis of pancreatic cancer patients according to claim 1, characterized in that: The cycle threshold is 24 months.

9. A system for evaluating the effect of B cells on the prognosis of pancreatic cancer patients, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 8 when executing the computer program.

Images