Gastric cancer tissue typing assisted recognition method based on digital pathological section

Abstract

The present application relates to the technical field of image processing, in particular to a gastric cancer tissue typing auxiliary identification method based on digital pathological sections. The method performs morphological feature identification on the edge information in the grayscale image of the section, and screens out normal cell edges and abnormal cell edges according to the size of the morphological feature index and the shape of the cell edge curve. According to the adjacent relationship, the first abnormal area and the second abnormal area are screened out, and then the enhancement coefficient is calculated and enhanced for the two kinds of abnormal areas. According to the enhanced gastric pathological image, the recognition result of gastric cancer tissue typing is obtained. The object of the enhancement process of the present application is limited to the abnormal area, and different degrees of enhancement are performed for different categories of abnormal areas, which reduces the calculation amount of the enhancement process, makes the enhancement process targeted, and does not cause excessive enhancement, obtains a clear section image for tissue typing identification, and improves the efficiency of tissue typing identification.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a gastric cancer tissue typing auxiliary identification method based on digital pathological sections. Background Art

[0002] In the process of using digital slice technology to perform typing and identification of gastric cancer tissue, the image quality may be affected by a variety of factors, such as errors in the slice preparation process, uneven staining, insufficient resolution of the imaging equipment, etc. These factors may cause image blur and loss of details, thereby affecting the identification effect of gastric cancer tissue typing. In the existing technology, the slice image is enhanced before typing and identification, thereby improving the image quality and ensuring the identification effect. However, the enhancement of the slice image in the existing technology is a global enhancement process. In fact, in addition to diseased cells, the image also includes normal cells. For the typing and identification process, the information of diseased cells is more important than that of normal cells. If the slice image is globally enhanced, the enhanced object will be non-targeted; and the degree of contrast between diseased cells and normal cells in the image is not taken into account, resulting in excessive enhancement of the information of diseased cells, increasing the computational complexity of the enhancement process, and affecting the efficiency of tissue typing and identification. Summary of the Invention

[0003] In order to solve the technical problem in the prior art of image enhancement of slice images, in which the unreasonable setting of the enhancement degree leads to an increase in the computational complexity of the enhancement process, thus affecting the efficiency of tissue typing and recognition, the present invention aims to provide a method for assisting the identification of gastric cancer tissue typing based on digital pathological sections. The technical solution adopted is as follows:

[0004] The present invention proposes a gastric cancer tissue typing auxiliary identification method based on digital pathological sections, the method comprising:

[0005] Obtaining a grayscale image of a digital pathological section of the stomach; obtaining a plurality of cell edge curves in the grayscale image;

[0006] For each cell edge curve, a morphological characteristic index of each cell edge curve is obtained according to the size of the area enclosed by the cell edge curve and the degree of edge smoothness; the cell edge curve is classified according to the morphological characteristic index to obtain multiple edge categories; and normal cell edges and abnormal cell edges are screened according to the size of the morphological characteristic index and the degree of edge smoothness in each edge category;

[0007] The abnormal cell edges adjacent to the normal cell edges are defined as first abnormal cell edges, and the other abnormal cell edges are defined as second abnormal cell edges; a plurality of first abnormal regions are defined based on the distances between the first abnormal cell edges; and a plurality of second abnormal regions are defined based on the distances between the second abnormal cell edges;

[0008] Obtaining an enhancement coefficient for each first abnormal region based on a grayscale difference between the first abnormal region and an adjacent normal cell region and the number of first abnormal cell edges within the first abnormal region and performing enhancement; obtaining an enhancement coefficient for each second abnormal region based on the number of second abnormal cell edges within the second abnormal region and performing enhancement; and obtaining an enhanced gastric pathological image;

[0009] The identification results of gastric cancer tissue typing were obtained based on the enhanced gastric pathological images.

[0010] Furthermore, the method for obtaining the cell edge curve includes:

[0011] An adaptive morphological edge detection algorithm is used to obtain edge pixel points in the grayscale image; each edge pixel point is used as a point to be grown. If other edge pixel points exist within a preset neighborhood range of the point to be grown, the other edge pixel points are used as new points to be grown and continue to grow until no more growth is possible, or the preset neighborhood range of the point to be grown contains the initial growth point, and the growth result is used as a cell edge curve to obtain all cell edge curves.

[0012] Furthermore, the method for obtaining the edge smoothness includes:

[0013] The tangent angle difference between adjacent edge pixel points on the cell edge curve is obtained, and the average tangent angle difference is negatively correlated with the difference to obtain the edge smoothness.

[0014] Furthermore, the method for obtaining the morphological characteristic index includes:

[0015] The area enclosed by the cell edge curve is taken as the cell area, the maximum distance between edge pixels on the cell edge curve is taken as the size length of the cell area, the product of the area and the size length of the cell area is taken as the size scale feature of the cell area, and the product of the size scale feature and the edge smoothness is normalized to obtain the morphological characteristic index.

[0016] Furthermore, screening out normal cell edges and abnormal cell edges includes:

[0017] If the morphological feature index in the edge category is within the preset index range, the edge category is used as the edge category to be selected; if the average edge smoothness in the edge category to be selected is greater than the preset threshold, the edge category to be selected is used as the category of normal cell edge composition, and the other edge categories are all categories of abnormal cell edge composition.

[0018] Furthermore, the method for screening the first abnormal cell edge includes:

[0019] Taking each normal cell edge as the center, a neighborhood judgment area of ​​each normal cell edge is constructed according to a preset neighborhood size, and the abnormal cell edge in the neighborhood judgment area is used as the first abnormal cell edge.

[0020] Furthermore, the method for dividing the first abnormal area and the second abnormal area includes:

[0021] Obtaining a neighborhood judgment region of each first abnormal cell edge, taking the first abnormal cell edges intersecting the neighborhood judgment regions as first abnormal cell edges of the same type, and the region formed by the first abnormal cell edges of the same type as the first abnormal region;

[0022] Obtain a neighborhood judgment region of each second abnormal cell edge, take the second abnormal cell edges intersecting the neighborhood judgment regions as similar second abnormal cell edges, and take the region formed by the similar second abnormal cell edges as the second abnormal region.

[0023] Furthermore, the method for obtaining the enhancement coefficient of the first abnormal area includes:

[0024] The area enclosed by the edges of abnormal cells is regarded as the abnormal cell area, and the area enclosed by the edges of normal cells is regarded as the normal cell area; for each first abnormal cell edge, the grayscale difference between the abnormal cell area enclosed by the first abnormal cell edge and the nearest normal cell area is obtained; the average grayscale difference in the first abnormal area is obtained, the ratio of the number of first abnormal cell edges in the first abnormal area to the average grayscale difference is normalized to obtain the initial enhancement coefficient of the first abnormal area, and the sum of the initial enhancement coefficient and the positive integer 1 is used as the enhancement coefficient of the first abnormal area.

[0025] Furthermore, the method for obtaining the enhancement coefficient of the second abnormal area includes:

[0026] Normalizing the number of second abnormal cell edges in the second abnormal region to obtain an initial enhancement coefficient of the second abnormal region, and taking the sum of the initial enhancement coefficient and the positive integer 1 as the enhancement coefficient of the second abnormal region.

[0027] Furthermore, any one of the first abnormal area and the second abnormal area is used as the area to be enhanced; the original pixel value of each pixel point in the area to be enhanced is multiplied by the enhancement coefficient to obtain the enhanced pixel value and the pixel value is replaced to obtain the enhanced area.

[0028] The present invention has the following beneficial effects:

[0029] The present invention first identifies the edge information in the grayscale image of the slice by morphological feature recognition, and selects the normal cell edges and abnormal cell edges based on the size of the morphological feature index and the morphology of the cell edge curve. Because there are obvious morphological differences between normal cells and abnormal cells, which are reflected in size changes and edge morphology, the morphological feature index and the degree of edge smoothness can effectively select the normal cell edges and abnormal cell edges. Based on the proximity relationship, the first abnormal area and the second abnormal area are selected, and then the enhancement coefficient is calculated and enhanced for each of the two abnormal areas. The object of the enhancement process is limited to the abnormal area, and different levels of enhancement are applied to different types of abnormal areas. This reduces the computational complexity of the enhancement process, makes the enhancement process targeted, and does not cause excessive enhancement. A clear slice image is obtained for tissue typing identification, thereby improving the efficiency of tissue typing identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0031] Figure 1 A flow chart of a method for assisting in the identification of gastric cancer tissue typing based on digital pathological sections provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0032] To further illustrate the technical means and effectiveness of the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of a method for assisting in the identification of gastric cancer tissue typing based on digital pathological sections proposed by the present invention. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0033] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0034] The following describes in detail a method for assisting in the identification of gastric cancer tissue typing based on digital pathological sections provided by the present invention with reference to the accompanying drawings.

[0035] See also Figure 1Fig. 1 shows a flow chart of a gastric cancer tissue typing auxiliary identification method based on digital pathological sections according to an embodiment of the present application, which comprises the following steps:

[0036] Step S1: obtaining a gray-scale image of a gastric digital pathological section; and obtaining a plurality of cell edge curves in the gray-scale image.

[0037] The embodiment of the present application aims to enhance the image before identifying the tissue typing of the section image of the gastric cancer tissue. Therefore, the image of the gastric digital pathological section is obtained and converted into a gray-scale image to facilitate subsequent morphological analysis. In the embodiment of the present application, the TDI-CCD linear scanning technology is used to convert the information on the slide into data section information quickly and comprehensively.

[0038] The gray-scale image contains cell information of the stomach, including normal cells and abnormal cells caused by gastric cancer. In order to enhance the image, it is necessary to distinguish between the diseased cells and the normal cells. The diseased cells and the normal cells have obvious differences in morphology, so the cell edge curves in the gray-scale image are extracted first, and then identified in the subsequent steps by using morphological features.

[0039] Preferably, in the embodiment of the present application, an adaptive morphological edge detection algorithm is used to obtain the edge pixel points in the gray-scale image; each edge pixel point is taken as a to-be-grown point, if there are other edge pixel points in the preset neighborhood range of the to-be-grown point, the other edge pixel points are taken as new to-be-grown points for continuous growth until no growth is possible, or the preset neighborhood range of the to-be-grown point contains the initial growth point, the growth result is taken as a cell edge curve, and all cell edge curves are obtained.

[0040] In an embodiment of the present application, the preset neighborhood range is set to an 8-neighborhood range. Further considering that other edge information in the image besides the cell edge may cause interference, in an embodiment of the present application, after obtaining the growth result, the length of the edge corresponding to the production result is used to remove the growth result with a smaller length. The length of the growth result with a smaller length can be removed by setting a threshold, clustering and classifying, etc., which are all well-known technical means to those skilled in the art, and will not be described and limited in detail.

[0041] Step S2: for each cell edge curve, obtaining a morphological feature index of each cell edge curve according to the size of the region surrounded by the cell edge curve and the edge smoothness; classifying the cell edge curves according to the morphological feature index to obtain a plurality of edge categories; and screening out normal cell edges and abnormal cell edges according to the size of the morphological feature index and the edge smoothness in each edge category.

[0042] For normal cells, the edges are regular, the edge boundaries between cells are clear, and the edge features are smooth and continuous. Due to the pathological reasons, the edges of abnormal cells appear irregular morphologies such as jagged, wavy or broken. For normal cells, the sizes are relatively consistent, and the cell region sizes are similar. Therefore, the properties of the cell edge curve can be classified based on the size of the cell region and the smoothness of the edge. Therefore, the embodiment of the present application first obtains the morphological feature index of each cell edge curve according to the size of the region surrounded by the cell edge curve and the smoothness of the edge. That is, the morphological feature index can represent the characteristics of the cell size and the edge smoothness, and further classify the cell edge curve according to the morphological feature index to obtain multiple edge categories. That is, the cell edge curves in one edge category are considered as the edges of the same kind of cells. According to the size of the morphological feature index of the edge category and the smoothness of the edge, the normal cell edge and the abnormal cell edge can be screened out.

[0043] Preferably, in an embodiment of the present application, the method for obtaining the edge smoothness comprises:

[0044] obtaining the tangent angle difference between adjacent edge pixels on the cell edge curve, performing negative correlation mapping on the average tangent angle difference, and obtaining the edge smoothness. It should be noted that the average tangent angle difference cannot be 0 because the cell edge cannot be a straight line. Therefore, in an embodiment of the present application, the reciprocal of the average tangent angle difference is taken as the edge smoothness.

[0045] Preferably, in an embodiment of the present application, the method for obtaining the morphological feature index comprises:

[0046] The region surrounded by the cell edge curve is taken as the cell region, the maximum distance between the edge pixels on the cell edge curve is taken as the size length of the cell region, the product of the area and the size length of the cell region is taken as the size scale feature of the cell region, and the product of the size scale feature and the edge smoothness is normalized to obtain the morphological feature index. That is, the larger the size scale feature and the edge smoothness, the larger the morphological feature index.

[0047] It should be noted that the cell edge curve may not be a closed edge, which cannot surround a closed region. Therefore, the curve fitting method is used to complete the edge to obtain the cell region. The curve fitting can use the least square method and other fitting methods, which are well known to those skilled in the art and will not be described here.

[0048] It should be noted that the normalization processing can be realized by using range standardization, function mapping method and other basic mathematical means, which are well known to those skilled in the art and will not be described here.

[0049] In order to analyze the cell edge curve and further reduce the calculation amount, the cell edge curve is classified according to the morphological feature index, and a plurality of edge categories are obtained. In the embodiment of the present application, considering that the morphological feature index is a normalized value between 0 and 1, the morphological feature index is divided into ten intervals with 0.1 as an interval, and each interval corresponds to an edge category, so as to complete the classification of the edge category.

[0050] Preferably, in an embodiment of the present application, the normal cell edge and the abnormal cell edge are screened, including:

[0051] If the morphological feature index in the edge category is within the preset index interval, the edge category is taken as a selected edge category; if the average edge smoothness in the selected edge category is greater than a preset threshold, the selected edge category is taken as a category composed of normal cell edges, and other edge categories are categories composed of abnormal cell edges.

[0052] It should be noted that the embodiment of the present application is aimed at image enhancement containing normal cells and abnormal cells at the same time, so there is no case that the image does not contain normal cells or abnormal cells. Therefore, the normal cell edge category and the abnormal cell edge category can be accurately identified by setting two conditions. The preset index interval and the preset threshold can be set according to the parameters of the camera and the actual type of the cell, for example, if the camera parameters cause the cells in the field of view to be small, the index interval should be set smaller, so it is not limited and described here.

[0053] Step S3: The abnormal cell edge adjacent to the normal cell edge is taken as a first abnormal cell edge, and other abnormal cell edges are taken as second abnormal cell edges; a plurality of first abnormal regions are divided according to the distance between the first abnormal cell edges; and a plurality of second abnormal regions are divided according to the distance between the second abnormal cell edges.

[0054] The embodiment of the present application considers that the abnormal cell adjacent to the normal cell has a relatively obvious contrast, so it does not need to be enhanced too much to highlight its image features; and for the abnormal cell far away from the normal cell, it can be normally enhanced. Therefore, the abnormal cells in the two states need to be analyzed respectively, so they need to be classified, and the abnormal cell edge adjacent to the normal cell edge is taken as a first abnormal cell edge, and other abnormal cell edges are taken as second abnormal cell edges.

[0055] For abnormal cells, because they are generated due to carcinogenesis, their morphology is relatively complex, and if there is an abnormal cell aggregation state, it will cause morphological overlap, leading to inaccurate information and affecting subsequent tissue typing recognition. Therefore, when performing image enhancement, the aggregation state of abnormal cells needs to be considered, that is, the more aggregated, the greater the degree of image enhancement required. Therefore, a plurality of first abnormal regions can be divided according to the distance between the edges of the first abnormal cells; a plurality of second abnormal regions can be divided according to the distance between the edges of the second abnormal cells.

[0056] Preferably, in an embodiment of the present application, the screening method of the first abnormal cell edge comprises:

[0057] With each normal cell edge as the center, a neighborhood judgment region of each normal cell edge is constructed according to a preset neighborhood size, and the abnormal cell edge in the neighborhood judgment region is regarded as the first abnormal cell edge. In an embodiment of the present application, the preset neighborhood size can be set as the size length of the central cell region, and it is considered that the neighborhood judgment region intersects with the abnormal cell edge, which satisfies the judgment condition of the first abnormal cell edge.

[0058] Further, the division method of the first abnormal region and the second abnormal region comprises:

[0059] The neighborhood judgment region of each first abnormal cell edge is obtained, and the first abnormal cell edges intersecting with the neighborhood judgment region are regarded as the same type of first abnormal cell edges, and the region composed of the same type of first abnormal cell edges is regarded as the first abnormal region.

[0060] The neighborhood judgment region of each second abnormal cell edge is obtained, and the second abnormal cell edges intersecting with the neighborhood judgment region are regarded as the same type of second abnormal cell edges, and the region composed of the same type of second abnormal cell edges is regarded as the second abnormal region.

[0061] Step S4: According to the gray difference between the first abnormal region and the adjacent normal cell edge, and the number of first abnormal cell edges in the first abnormal region, the enhancement coefficient of each first abnormal region is obtained and enhanced; according to the number of second abnormal cell edges in the second abnormal region, the enhancement coefficient of each second abnormal region is obtained and enhanced; and the enhanced stomach pathological image is obtained.

[0062] As described in step S3, for the first abnormal region, it is necessary to consider the adjacent normal cells when enhancing it. The greater the grayscale difference between the first abnormal region and the adjacent normal cell region, the less significant enhancement is required for the first abnormal region, as it already has a clear contrast with the normal cells. The first abnormal region is similar to the second abnormal region, and both need to consider the number of abnormal cells represented by the abnormal cell edges in the region. That is, the greater the number of abnormal cell edges, the more aggregated the abnormal cells in the abnormal region are, and the greater the degree of enhancement is required. Therefore, based on the grayscale difference between the first abnormal region and the adjacent normal cell region, and the number of first abnormal cell edges in the first abnormal region, the enhancement coefficient of each first abnormal region is obtained and enhanced; based on the number of second abnormal cell edges in the second abnormal region, the enhancement coefficient of each second abnormal region is obtained and enhanced; and the enhanced gastric pathology image is obtained.

[0063] Preferably, in one embodiment of the present invention, the method for obtaining the enhancement coefficient of the first abnormal region includes:

[0064] The area enclosed by the edges of abnormal cells is considered the abnormal cell area, and the area enclosed by the edges of normal cells is considered the normal cell area. For each first abnormal cell edge, the grayscale difference between the abnormal cell area enclosed by the first abnormal cell edge and the nearest normal cell area is obtained. It should be noted that the nearest normal cell area is the closest normal cell area. The distance between cell areas can be represented by the distance between the area centroids, and the grayscale difference is the absolute difference between the average grayscale values ​​within the area.

[0065] Each abnormal cell region in the first abnormal region corresponds to a grayscale difference. Therefore, the average grayscale difference in the first abnormal region is obtained. The ratio of the number of first abnormal cell edges in the first abnormal region to the average grayscale difference is normalized to obtain the initial enhancement coefficient of the first abnormal region. Because the initial enhancement coefficient is a value between 0 and 1, in this embodiment of the present invention, the sum of the initial enhancement coefficient and the positive integer 1 is used as the enhancement coefficient of the first abnormal region. That is, the enhancement coefficient is an enhancement multiple, and its value range is 1 to 2.

[0066] Preferably, in one embodiment of the present invention, the method for obtaining the enhancement coefficient of the second abnormal region includes:

[0067] The number of second abnormal cell edges in the second abnormal region is normalized to obtain an initial enhancement coefficient of the second abnormal region, and the sum of the initial enhancement coefficient and the positive integer 1 is used as the enhancement coefficient of the second abnormal region.

[0068] Preferably, in one embodiment of the present invention, a linear enhancement method is used to enhance the pixel values ​​in the abnormal area, specifically: any one of the first abnormal area and the second abnormal area is used as the area to be enhanced; the original pixel value of each pixel point in the area to be enhanced is multiplied by the enhancement coefficient to obtain the enhanced pixel value and the pixel value is replaced to obtain the enhanced area.

[0069] Step S5: Obtaining a recognition result of gastric cancer tissue typing based on the enhanced gastric pathological image.

[0070] In the embodiment of the present invention, gastric cancer tissue typing and identification are performed based on a trained recognition model, which can be regarded as a prior art and will not be described in detail here.

[0071] In summary, the embodiment of the present invention performs morphological feature recognition on the edge information in the grayscale image of the slice, and selects the normal cell edge and the abnormal cell edge according to the size of the morphological feature index and the morphology of the cell edge curve. The first abnormal area and the second abnormal area are selected based on the proximity relationship, and then the enhancement coefficient is calculated and enhanced for the two abnormal areas respectively, and the identification result of the gastric cancer tissue typing is obtained based on the enhanced gastric pathology image. The object of the enhancement process of the present invention is limited to the abnormal area, and different degrees of enhancement are performed for different categories of abnormal areas, which reduces the amount of calculation of the enhancement process, makes the enhancement process targeted, and does not cause excessive enhancement, thereby obtaining a clear slice image for tissue typing identification, thereby improving the efficiency of tissue typing identification.

[0072] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0073] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A gastric cancer tissue typing auxiliary identification method based on digital pathological sections, characterized in that: The method comprises: Obtaining a grayscale image of a digital pathological section of the stomach; obtaining a plurality of cell edge curves in the grayscale image; For each cell edge curve, a morphological characteristic index of each cell edge curve is obtained according to the size of the area enclosed by the cell edge curve and the degree of edge smoothness; the cell edge curve is classified according to the morphological characteristic index to obtain multiple edge categories; and normal cell edges and abnormal cell edges are screened according to the size of the morphological characteristic index and the degree of edge smoothness in each edge category; The abnormal cell edges adjacent to the normal cell edges are defined as first abnormal cell edges, and the other abnormal cell edges are defined as second abnormal cell edges; a plurality of first abnormal regions are defined based on the distances between the first abnormal cell edges; and a plurality of second abnormal regions are defined based on the distances between the second abnormal cell edges; Obtaining an enhancement coefficient for each first abnormal region based on a grayscale difference between the first abnormal region and an adjacent normal cell region and the number of first abnormal cell edges within the first abnormal region and performing enhancement; obtaining an enhancement coefficient for each second abnormal region based on the number of second abnormal cell edges within the second abnormal region and performing enhancement; and obtaining an enhanced gastric pathological image; Obtain identification results of gastric cancer tissue typing based on enhanced gastric pathological images; The method for obtaining the enhancement coefficient of the first abnormal area includes: The area enclosed by the edges of abnormal cells is defined as the abnormal cell area, and the area enclosed by the edges of normal cells is defined as the normal cell area. For each first abnormal cell edge, a grayscale difference is obtained between the abnormal cell area enclosed by the first abnormal cell edge and the nearest normal cell area. An average grayscale difference is obtained in the first abnormal area, and the ratio of the number of first abnormal cell edges in the first abnormal area to the average grayscale difference is normalized to obtain an initial enhancement coefficient for the first abnormal area. The sum of the initial enhancement coefficient and the positive integer 1 is used as the enhancement coefficient for the first abnormal area. The method for obtaining the enhancement coefficient of the second abnormal area includes: Normalizing the number of second abnormal cell edges in the second abnormal region to obtain an initial enhancement coefficient of the second abnormal region, and taking the sum of the initial enhancement coefficient and the positive integer 1 as the enhancement coefficient of the second abnormal region.

2. The method for assisting in the identification of gastric cancer tissue typing based on digital pathological sections according to claim 1, characterized in that: The method for obtaining the cell edge curve includes: An adaptive morphological edge detection algorithm is used to obtain edge pixel points in the grayscale image; each edge pixel point is used as a point to be grown. If other edge pixel points exist within a preset neighborhood range of the point to be grown, the other edge pixel points are used as new points to be grown and continue to grow until no more growth is possible, or the preset neighborhood range of the point to be grown contains the initial growth point, and the growth result is used as a cell edge curve to obtain all cell edge curves.

3. The method for assisting in the identification of gastric cancer tissue typing based on digital pathological sections according to claim 1, characterized in that: The method for obtaining the edge smoothness includes: The tangent angle difference between adjacent edge pixel points on the cell edge curve is obtained, and the average tangent angle difference is negatively correlated with the difference to obtain the edge smoothness.

4. The method for assisting in the identification of gastric cancer tissue typing based on digital pathological sections according to claim 1, characterized in that: Methods for obtaining morphological characteristic indicators include: The area enclosed by the cell edge curve is taken as the cell area, the maximum distance between edge pixels on the cell edge curve is taken as the size length of the cell area, the product of the area and the size length of the cell area is taken as the size scale feature of the cell area, and the product of the size scale feature and the edge smoothness is normalized to obtain the morphological characteristic index.

5. The method for assisting in the identification of gastric cancer tissue typing based on digital pathological sections according to claim 1, characterized in that: The screening of normal cell edges and abnormal cell edges includes: If the morphological feature index in the edge category is within the preset index range, the edge category is used as the edge category to be selected; if the average edge smoothness in the edge category to be selected is greater than the preset threshold, the edge category to be selected is used as the category of normal cell edge composition, and the other edge categories are all categories of abnormal cell edge composition.

6. The method for assisting in the identification of gastric cancer tissue typing based on digital pathological sections according to claim 1, characterized in that: The method for screening the first abnormal cell edge comprises: Taking each normal cell edge as the center, a neighborhood judgment area of ​​each normal cell edge is constructed according to a preset neighborhood size, and the abnormal cell edge in the neighborhood judgment area is used as the first abnormal cell edge.

7. The method for assisting gastric cancer tissue typing and identification based on digital pathological sections according to claim 6, characterized in that: The method for dividing the first abnormal area and the second abnormal area includes: Obtaining a neighborhood judgment region of each first abnormal cell edge, taking the first abnormal cell edges intersecting the neighborhood judgment regions as first abnormal cell edges of the same type, and the region formed by the first abnormal cell edges of the same type as the first abnormal region; Obtain a neighborhood judgment region of each second abnormal cell edge, take the second abnormal cell edges intersecting the neighborhood judgment regions as similar second abnormal cell edges, and take the region formed by the similar second abnormal cell edges as the second abnormal region.

8. The method for assisting gastric cancer tissue typing based on digital pathological sections according to claim 1, characterized in that: Taking any one of the first abnormal area and the second abnormal area as the area to be enhanced; The original pixel value of each pixel point in the area to be enhanced is multiplied by the enhancement coefficient to obtain an enhanced pixel value, and the pixel value is replaced to obtain the enhanced area.

Images