A method and device for identifying villus on the surface of colon mucosa in inflammatory bowel disease

Abstract

The present invention provides a method and device for identifying the villus on the surface of colon mucosa in inflammatory bowel disease, and relates to the field of image data processing technology. The method includes: based on the grayscale image of the pathological section staining image of colon biopsy tissue, determining the first possibility that each pixel point in the pathological section staining image is located in the background area; calculating the second possibility that each pixel point is located in the low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image; adjusting the size of the structural element in the process of operating the pathological section staining image based on the second possibility; and identifying the degree of villus on the surface of colon mucosa according to the pathological section staining image after the size adjustment. The present invention adjusts the size of the structural element in the low-quality fuzzy area to effectively remove the low-quality fuzzy area, and more accurately identifies the degree of villus on the surface of colon mucosa in inflammatory bowel disease.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and in particular to a method and device for identifying villus on the surface of colon mucosa in inflammatory bowel disease. Background Art

[0002] The degree of villi on the colonic mucosal surface can indicate the presence of inflammatory bowel disease.

[0003] Normally, pathological sections of the colon mucosa are taken, and the H&E (Hematoxylin and Eosin) staining method is usually used to stain the pathological sections of the colon biopsy tissue, and then the slides are scanned by a scanner to obtain pathological images of the colon biopsy tissue, so that the collected pathological sections of the colon biopsy tissue can show clear tissue structure contrast. This contrast can assist in observing the cell morphology and structure in the pathological sections of the colon biopsy tissue.

[0004] On this basis, a segmentation model can be trained to identify the villous part of the colon mucosa in pathological sections, which can replace manual judgment of the degree of villization on the surface of the colon mucosa, thereby reducing the workload of doctors.

[0005] However, in the process of obtaining pathological images using a scanner, due to the limitations of the optical imaging system, such as photon noise and lens aberrations, low-quality and blurred areas will exist in the obtained pathological images, thereby affecting the recognition of the biopsy tissue area in the pathological image, and further affecting the extraction of the villi features of the colon mucosal surface. Summary of the invention

[0006] The present invention provides a method and device for identifying villus on the surface of colon mucosa in inflammatory bowel disease, so as to solve the defect in the prior art that low-quality fuzzy areas in pathological images affect the identification of villus on the surface of colon mucosa, and realize a more accurate method and device for identifying villus on the surface of colon mucosa in inflammatory bowel disease.

[0007] The present invention provides a method for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease, comprising:

[0008] Based on the grayscale image of the pathological section staining image of the colon biopsy tissue, determining a first possibility that each pixel point in the pathological section staining image is located in the background area;

[0009] Calculating a second possibility that each pixel point is located in a low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image;

[0010] adjusting the size of the structural element during the operation of the pathological section staining image based on the second possibility;

[0011] The degree of villi on the surface of the colon mucosa is identified based on the size-adjusted staining image of the pathological section.

[0012] According to a method for identifying villus on the surface of colon mucosa in inflammatory bowel disease provided by the present invention, the step of determining the first possibility that each pixel point in the pathological section staining image is located in the background area based on the grayscale image of the pathological section staining image of the colon biopsy tissue specifically includes:

[0013] Calculate the first pixel mean and the second pixel mean of the row and the column to which each pixel point in the grayscale image belongs respectively;

[0014] The larger one of the first pixel mean and the second pixel mean of each pixel in the grayscale image after normalization is used as the first possibility of each pixel in the pathological section staining image.

[0015] According to a method for identifying villus on the surface of colon mucosa in inflammatory bowel disease provided by the present invention, the step of determining the first possibility that each pixel point in the pathological section staining image is located in the background area based on the grayscale image of the pathological section staining image of the colon biopsy tissue specifically includes:

[0016] Calculate the first pixel mean and the second pixel mean of the row and the column to which each pixel point in the grayscale image belongs respectively;

[0017] The larger one of the first pixel mean and the second pixel mean of each pixel in the grayscale image is normalized as the third possibility of each pixel in the grayscale image;

[0018] Calculate the ratio of the pixel value of each pixel in the grayscale image to 255 as the first weight;

[0019] The product of the third possibility of each pixel in the grayscale image and the first weight is used as the first possibility of each pixel in the pathological section staining image.

[0020] According to a method for identifying villus on the surface of colon mucosa in inflammatory bowel disease provided by the present invention, the step of calculating the second possibility that each pixel point is located in the low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image specifically includes:

[0021] Acquire an edge image of a staining image of a pathological section of the colon biopsy tissue;

[0022] For each pixel point in the edge image, determining a first variance of pixels within a preset range centered on each pixel point;

[0023] Calculating the second variance of all pixels in the edge image;

[0024] The second possibility of each pixel point in the pathological section staining image is determined according to the ratio of the first variance to the second variance and the first possibility.

[0025] According to a method for identifying villus on the surface of colon mucosa in inflammatory bowel disease provided by the present invention, the step of calculating the second possibility that each pixel point is located in the low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image further includes:

[0026] For each pixel point in the pathological section staining image, when the first possibility of the pixel point is not less than a first preset threshold, the second possibility of the pixel point is increased.

[0027] According to a method for identifying villus on the surface of colon mucosa in inflammatory bowel disease provided by the present invention, the step of calculating the second possibility that each pixel point is located in the low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image further includes:

[0028] Calculate a first average value of the number of edge pixels within all the preset ranges;

[0029] For each pixel point in the pathological section staining image, when the number of edge pixels within the preset range centered at the pixel point is less than the first mean value, the second possibility of the pixel point is increased.

[0030] According to a method for identifying villus on the surface of colon mucosa in inflammatory bowel disease provided by the present invention, the step of adjusting the size of the structural element in the process of operating the pathological section staining image based on the second possibility specifically includes:

[0031] When the second possibility of the pixel point is not less than a second preset threshold, the size of the structure element corresponding to the pixel point is increased.

[0032] The present invention also provides a method for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease, comprising:

[0033] A first determination module is used to determine a first possibility that each pixel point in the pathological section staining image is located in the background area based on the grayscale image of the pathological section staining image of the colon biopsy tissue;

[0034] A second determination module is used to calculate a second possibility that each pixel point is located in a low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image;

[0035] An adjustment module, configured to adjust the size of the structural element in the process of operating the pathological section staining image based on the second possibility;

[0036] The recognition module is used to recognize the degree of villi on the surface of the colon mucosa according to the size-adjusted pathological section staining image.

[0037] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a method for identifying villus formation on the surface of the colon mucosa in inflammatory bowel disease as described in any one of the above methods is implemented.

[0038] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease as described in any one of the above methods.

[0039] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease as described in any one of the above methods.

[0040] The present invention provides a method and device for identifying villus on the surface of colon mucosa in inflammatory bowel disease. By processing the staining image of pathological sections of colon biopsy tissue, the structural element can have a larger size when it is located in a low-quality fuzzy area, so as to effectively remove the low-quality fuzzy area in the image. The structural element can have a smaller size when it is located in the area of ​​colon biopsy tissue in the image. While effectively removing the low-quality fuzzy area in the staining image, the image details of the colon biopsy tissue in the staining image can also be retained, which is beneficial to the feature extraction of the colon biopsy tissue in the image, thereby improving the recognition accuracy of villus on the surface of colon mucosa in inflammatory bowel disease. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0042] Figure 1 It is a schematic flow chart of a method for identifying villi on the surface of colon mucosa in inflammatory bowel disease provided by the present invention;

[0043] Figure 2 It is a schematic diagram mainly used to display low-quality fuzzy areas in a method for identifying villi on the surface of colon mucosa in inflammatory bowel disease provided by the invention;

[0044] Figure 3It is a schematic diagram of the structure of a device for identifying villi on the surface of colon mucosa of inflammatory bowel disease provided by the present invention;

[0045] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] Combine the following Figure 1 The present invention introduces a method for identifying villi on the surface of colon mucosa in inflammatory bowel disease, such as Figure 1 As shown, including:

[0048] Step 101, based on the grayscale image of the pathological section staining image of the colon biopsy tissue, determining a first possibility that each pixel point in the pathological section staining image is located in the background area;

[0049] First, the pathological sections of the colon biopsy tissue are stained by H&E staining, and images of the pathological sections are acquired using a scanner capable of providing high-resolution images to obtain H&E staining images of the pathological sections of the colon biopsy tissue, which are collectively referred to as pathological section staining images in the following text.

[0050] It is understandable that the background area represents the location where the colon biopsy tissue does not appear in the pathological section staining image. In the grayscale image of the pathological section staining image, the background area is usually brighter than the area where the colon biopsy tissue appears, so in the grayscale image, the pixel value of the pixel point located in the background area is usually higher.

[0051] At the same time, limitations of the optical imaging system produce low-quality blurred areas such as Figure 2 The area encircled by the middle dotted circle is generally a silky color block free around the colon biopsy tissue, whose gray value is lower than that of the background area, but lacks texture details.

[0052] On this basis, in order to reduce the impact of low-quality blurred areas in the pathological section staining image, it is necessary to make a preliminary judgment on whether each position in the image is in the staining area based on the grayscale value of the pixel point at each position in the image, and determine the first possibility that each pixel point in the pathological section staining image is located in the background area.

[0053] The higher the first possibility, the more likely the corresponding pixel point is to be the pixel point corresponding to the background area, and the smaller the first possibility, the more likely it is to be the pixel point corresponding to the biopsy tissue.

[0054] Step 102, calculating a second possibility that each pixel point is located in a low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image;

[0055] Furthermore, it can be understood that the biopsy tissue area in the pathological section staining image has richer texture information than the low-quality blurred area, that is, the different texture features of the colon biopsy tissue area and the low-quality blurred area in the grayscale image. Therefore, based on the pixel points that are more likely to be non-background areas determined based on the first possibility, it is possible to further identify whether the pixel points belong to the colon biopsy tissue area or the low-quality blurred area by judging their texture information.

[0056] Optionally, an edge image of the pathological section staining image is obtained by edge detection technology, and the richness of texture information in the area around each pixel is determined based on the edge image, so as to determine the second possibility that each pixel in the pathological section staining image is located in a low-quality blurred area based on the first possibility.

[0057] The greater the second possibility is, the greater the possibility that the pixel is located in a low-quality blurred area.

[0058] Step 103, adjusting the size of the structural element in the process of operating the pathological section staining image based on the second possibility;

[0059] It can be understood that by calculating the second possibility of each pixel in the pathological section staining image, that is, by screening the grayscale value of each pixel and the richness of the surrounding texture, pixels with lower grayscale values ​​and richer surrounding textures are obtained, and they are confirmed as pixels corresponding to the colon biopsy tissue area, and other pixels are confirmed as pixels corresponding to the background area and low-quality blurred areas.

[0060] On this basis, in the process of opening the pathological section staining image, for the pixel points corresponding to the biopsy tissue area, their structural elements are not processed; for the pixel points corresponding to the background area and the low-quality blurred area, their structural elements are enlarged for opening operation so as to remove them.

[0061] Step 104, identifying the degree of villi on the surface of the colon mucosa according to the size-adjusted pathological section staining image.

[0062] Each channel of the pathological section staining image is processed, and the processed channels are recombined into a color image to obtain the colon biopsy tissue area in the pathological section staining image and identify its villiformization degree.

[0063] At this time, the pixels corresponding to the background area and the low-quality blurred area are removed from the pathological section staining image after the opening operation, which to a certain extent avoids the process of identifying the degree of villus formation on the colon mucosal surface from being affected by the low-quality blurred area, thereby obtaining a more accurate recognition result of the degree of villus formation on the colon mucosal surface.

[0064] The present invention processes the pathological section staining image of the colon biopsy tissue, so that the structure element can have a larger size when it is located in the low-quality blurred area, so as to effectively remove the low-quality blurred area in the image, and can have a smaller size when the structure element is located in the area of ​​the colon biopsy tissue in the image. While effectively removing the low-quality blurred area in the staining image, the image details of the colon biopsy tissue in the staining image can also be retained, which is beneficial to the feature extraction of the colon biopsy tissue in the image, thereby improving the recognition accuracy of the villi on the surface of the colon mucosa in inflammatory bowel disease.

[0065] In a method for identifying villus on the surface of colon mucosa in inflammatory bowel disease of the present invention, the step of determining the first possibility that each pixel point in the pathological section staining image is located in the background area based on the grayscale image of the pathological section staining image of the colon biopsy tissue specifically includes:

[0066] Calculate the first pixel mean and the second pixel mean of the row and the column to which each pixel point in the grayscale image belongs respectively;

[0067] By comparing the pixel value of each pixel in the grayscale image with the average pixel value of its row and column, it is determined whether the pixel is more likely to correspond to the background area or the non-background area.

[0068] That is, the average pixel value of the grayscale image is represented by the first pixel mean and the second pixel mean.

[0069] The larger one of the first pixel mean and the second pixel mean of each pixel in the grayscale image after normalization is used as the first possibility of each pixel in the pathological section staining image.

[0070] Optionally, the first pixel mean and the second pixel mean are normalized by using a maximum and minimum value normalization method, so that they can be used as the first possibility for subsequent calculation.

[0071] In a specific embodiment, the first possibility As shown below:

[0072] ;

[0073] In the formula, Indicates that it is located at i The first pixel mean corresponding to the pixel point of the row, Represents the minimum value of the first pixel means corresponding to each row of the grayscale image; Represents the maximum value of the first pixel mean corresponding to each row of the grayscale image; W Indicates the length of the grayscale image.

[0074] Indicates that it is located at j The second pixel mean corresponding to the pixel points in the column, Represents the minimum value of the second pixel means corresponding to each column of the grayscale image; Represents the maximum value of the second pixel means corresponding to each column of the grayscale image; H Indicates the width of the grayscale image. represents the maximum value function, Shows the minimum function.

[0075] That is to say, the first pixel mean and the second pixel mean corresponding to each pixel point are normalized by maximum and minimum value normalization. When the maximum value of the two values ​​after normalization is smaller, it means that the corresponding pixel point is more likely to be a pixel point in the colon biopsy tissue area.

[0076] It should be noted that there are often tissue structures such as crypt structures in the biopsy tissue of the colon. In the grayscale image corresponding to the pathological section staining image of the colon biopsy tissue, there will also be areas with high grayscale values ​​in some positions of the crypt tissue, which will result in some tissue structures such as the colon biopsy tissue in the staining image having a higher initial probability of being in the background area.

[0077] The first pixel mean and the second pixel mean corresponding to each pixel in the grayscale image can, to a certain extent, measure whether the pixel is in the colon biopsy tissue area from the perspective of the entire image.

[0078] The higher the mean value of the first pixel corresponding to a row in the staining image, the more likely it is that there are pixels corresponding to the background area in this row of the image. The lower the mean value of the first pixel corresponding to a row in the staining image, the more likely it is that there are pixels corresponding to the colon biopsy tissue area in this row of the image.

[0079] The higher the second pixel mean corresponding to a column in the staining image, the more likely it is that there are pixels corresponding to the background area in this column in the image. The lower the second pixel mean corresponding to a column in the staining image, the more likely it is that there are pixels corresponding to the colon biopsy tissue area in this column in the image.

[0080] In a method for identifying villus on the surface of colon mucosa in inflammatory bowel disease of the present invention, the step of determining the first possibility that each pixel point in the pathological section staining image is located in the background area based on the grayscale image of the pathological section staining image of the colon biopsy tissue specifically includes:

[0081] Calculate the first pixel mean and the second pixel mean of the row and the column to which each pixel point in the grayscale image belongs respectively;

[0082] The larger one of the first pixel mean and the second pixel mean of each pixel in the grayscale image is normalized as the third possibility of each pixel in the grayscale image;

[0083] Calculate the ratio of the pixel value of each pixel in the grayscale image to 255 as the first weight;

[0084] The product of the third possibility of each pixel in the grayscale image and the first weight is used as the first possibility of each pixel in the pathological section staining image.

[0085] The method of calculating the third possibility is the same as the method of calculating the first possibility in the previous text, so it is not repeated here. At this time, the third possibility of the pixel point represents the level of the gray value of the pixel point in the overall gray value of the gray image.

[0086] On this basis, since the color of the background area in the pathological section staining image is usually bright white, and the pixel with a larger grayscale value is more likely to correspond to the bright white background area, the first possibility of the pixel can be obtained by weighting the calculated third possibility using the ratio of the pixel value of each pixel in the grayscale image to 255 as the first weight.

[0087] In a specific embodiment, the first possibility As shown below:

[0088] ;

[0089] In the formula, Indicates that it is located at i Row, No. j The pixel value of the pixel in the column.

[0090] It is understandable that when The larger the value of is, the greater the first possibility that the pixel is a pixel in the background area; The smaller the value, the more likely the pixel is to be the pixel corresponding to the colon biopsy tissue.

[0091] The present invention determines the first possibility of each pixel point in the pathological section staining image by combining the pixel value size of the pixel point itself and the grayscale degree of the pixel value of the pixel point in the grayscale image, so as to make the judgment result more accurate.

[0092] In a method for identifying villus on the surface of colon mucosa in inflammatory bowel disease of the present invention, the step of calculating the second possibility that each pixel point is located in the low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image specifically includes:

[0093] Acquire an edge image of a staining image of a pathological section of the colon biopsy tissue;

[0094] Furthermore, since the grayscale value of the low-quality blurred area is lower than that of the background area as a whole, the pixel points corresponding to the low-quality blurred area are easily identified as the pixel points that may correspond to the colon biopsy tissue area in the judgment based on the first possibility.

[0095] Therefore, on the basis of determining the first possibility of each pixel point, the second possibility of each pixel point can also be determined in combination with the texture richness of the pixels in the neighborhood of each pixel point, as the possibility that the pixel point is located in a low-quality blurred area.

[0096] Optionally, an edge map of the pathological section staining image is obtained by a Canny edge detection algorithm, wherein the edge map is composed of pixel points with pixel values ​​of 0 and 1, the edge in the staining image is 1 at the corresponding position in the edge map, and the remaining positions in the edge map are 0. The colon biopsy tissue area has more edges in the edge map. When there are more edge pixels around a pixel point, that is, there are more edges around a pixel point, the pixel point is more likely to be a pixel point corresponding to the colon biopsy tissue area.

[0097] That is, the number of edge pixels in the neighborhood of each pixel is used to characterize the texture features of the location where the pixel belongs.

[0098] For each pixel point in the edge image, determining a first variance of pixels within a preset range centered on each pixel point;

[0099] Calculating the second variance of all pixels in the edge image;

[0100] The second possibility of each pixel point in the pathological section staining image is determined according to the ratio of the first variance to the second variance and the first possibility.

[0101] To calculate the second likelihood of each pixel, we first set a K × KThe sliding window is centered on each pixel in the edge image, with a step size of 1, and slides gradually from left to right and from top to bottom, so that the preset range centered on each pixel is represented by the sliding window.

[0102] In one possible implementation, K is 31, that is, the pixels within the preset range are all pixels within a range of 31×31 centered on the currently calculated pixel. In other feasible implementations, the size of the sliding window can also be adjusted according to the pathological section staining image of the collected colon biopsy tissue and the magnification of the image acquisition device.

[0103] On this basis, the second possibility of each pixel As shown below:

[0104] ;

[0105] In the formula, Indicates that it is located at i Row, No. j The second possibility of the pixel point of the column; Indicates that the center of the sliding window moves to the i Row, No. j When the pixel points of the column are, the variance of all the pixels in the sliding window is the first variance; Represents the variance of all pixels in the edge image, which is the second variance; Represents an exponential function with the natural constant e as the base.

[0106] Specifically, The larger the value of is, the more likely the edge image area corresponding to the current sliding window is to have a complex texture structure. i Row, No. j The smaller the value, the more likely the pixel point in the column is to be in the colon biopsy tissue area; on the contrary, the smaller the value, the more likely the area corresponding to the window area in the staining image is to have a simple texture structure or even a textureless structure. i Row, No. j The pixels in the lower column are more likely to be in the low-quality blur area or background area.

[0107] In the method for identifying villus on the surface of colon mucosa in inflammatory bowel disease of the present invention, the step of calculating the second possibility that each pixel point is located in the low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image also includes:

[0108] For each pixel point in the pathological section staining image, when the first possibility of the pixel point is not less than a first preset threshold, the second possibility of the pixel point is increased.

[0109] Furthermore, the second possibility corresponding to each pixel point can be further corrected by setting a first preset threshold.

[0110] Specifically, as follows:

[0111] ;

[0112] In the above formula, the first preset threshold is set to 0.5.

[0113] That is to say, for any pixel point in the pathological section staining image, if its first possibility is not less than 0.5, it is considered to be a pixel point that is more inclined to the background area, so its second possibility is increased and it is corrected to a pixel point that is more inclined to the low-quality blurred area or the background area.

[0114] When the first possibility is less than 0.5, it is considered that the pixel point is more inclined to the colon biopsy tissue area, and thus the second possibility is not corrected.

[0115] Optionally, the first preset threshold is determined based on the pixel values ​​of the background area pixels. It is understandable that the pixel values ​​of the pixels located in the background area are higher than the pixel values ​​of the pixels in other areas, and the values ​​are more similar. Therefore, the pixels in the grayscale image that meet the above characteristics are determined as background area pixels, and the first preset threshold is determined based on the first possible value corresponding to such pixels.

[0116] In the method for identifying villus on the surface of colon mucosa in inflammatory bowel disease of the present invention, the step of calculating the second possibility that each pixel point is located in the low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image further includes:

[0117] Calculate a first average value of the number of edge pixels within all the preset ranges;

[0118] For each pixel point in the pathological section staining image, when the number of edge pixels within the preset range centered at the pixel point is less than the first mean value, the second possibility of the pixel point is increased.

[0119] Furthermore, the number of edge pixels within a preset range around a pixel point can be used as an influencing factor to modify the second possibility corresponding to each pixel point. Specifically, the more edge pixels there are within a preset range around a pixel point, the more likely the pixel point is to be a pixel point in the colon biopsy tissue area.

[0120] Optionally, the preset range around the pixel point can be represented by an area corresponding to the sliding window.

[0121] Specifically, as follows:

[0122] ;

[0123] In the formula, Indicates that it is located at i Row, No. j The number of all edge pixels in the sliding window corresponding to the pixel points in the column; Indicates the number of pixels in the edge image; Represents the mean of the number of all edge pixels in the sliding window corresponding to all pixels in the edge image.

[0124] That is to say, In the case of i Row, No. j The pixel point of the column is located in an area with fewer edges, that is, the pixel point is more likely to be a pixel point in the background area, thereby increasing its second possibility and correcting it to be a pixel point that is more likely to be a low-quality blurred area or a background area.

[0125] exist In the case of i Row, No. j The pixel point of the column is located in an area with more edges, that is, the pixel point is more likely to be a pixel point in the colon biopsy tissue area, so the second possibility is not corrected.

[0126] In addition, optionally, in the present application, the second possibility of each pixel point is calculated by setting a sliding window. Initially, the step size of the sliding window is set to 1. In order to improve the processing speed, when the sliding window slides on the edge image, if the number of all edge pixels in the sliding window is is 0, indicating that the sliding window is in the background area at this time, the sliding step size of the sliding window is adjusted to the window size K, and the first possibility of all pixels in the sliding window at this time is set to 1; when the number of all edge pixels in the sliding window When it is not 0, the step size of the window sliding is readjusted to 1.

[0127] In a method for identifying villus on the surface of colon mucosa in inflammatory bowel disease of the present invention, the step of adjusting the size of the structural element in the process of operating the pathological section staining image based on the second possibility specifically includes:

[0128] When the second possibility of the pixel point is not less than a second preset threshold, the size of the structure element corresponding to the pixel point is increased.

[0129] When performing an opening operation, the choice of the structure element affects whether the opening operation can effectively remove the low-quality blurred areas in the stained image. A small structure element can retain the details of the image, but cannot effectively remove the low-quality blurred areas in the stained image. Although a large structure element can effectively remove the low-quality blurred areas in the stained image, it will cause the loss of details and edge information in the image. The size of the structure element is adjusted based on the second possibility of each pixel in the following way.

[0130] The opening operation includes two processing steps: corrosion and expansion.

[0131] When the second likelihood of the pixel point is greater than or equal to the second preset threshold, a larger structure element is used to remove the low-quality blur area; when the second likelihood of the pixel point is less than the second preset threshold, the size of the structure element is set to 1, that is, no corrosion operation is performed;

[0132] In response to the size of the eroded structure element not being 1, the size of the expanded structure element is twice the size of the eroded structure element, otherwise, no expansion operation is performed.

[0133] Specifically, the size of the structural element satisfies the following relationship:

[0134] ;

[0135] ;

[0136] In the formula, Indicates the pathological section staining map located at i Row, No. j The size of the structural element of the erosion step for opening the pixels of the column; K Indicates the single-side size of the sliding window; Indicates that it is located at i Row, No. j The Euclidean distance between the 𝑘th pixel in the sliding window corresponding to the pixel point of the column and the center point of the window; Indicates the pathological section staining map located at i Row, No. j The size of the structural element in the dilation step of the pixel points in the column.

[0137] Optionally, the second preset threshold is set to 0.6 based on experience.

[0138] When the second possibility of the pixel point is not less than the second preset threshold, the pixel point is considered to be a pixel point in a non-colon biopsy tissue area, and the size of the structure element needs to be adjusted to facilitate its removal. Represents the maximum Euclidean distance between the pixel point in the sliding window and the center point of the sliding window. For the convenience of calculation, right Normalize it, Indicates that the center of the window slides to the edge image at the i Row, No. j When the pixel point of the column is k The relative distance between the pixel point and the center point of the sliding window; It represents the average value of the relative distance between all pixels in the sliding window and the center point of the sliding window. The larger the value, the larger the range of the low-quality fuzzy area in the sliding window. A larger structural element should be used for the pathological section staining image. i Row, No. j The smaller the value, the smaller the range of the low-quality fuzzy area in the sliding window. The smaller the structure element should be used for the first column of the pathological section staining image. i Row, No. j The pixel points of the column are opened.

[0139] in, It indicates the basic size of the structural element. The size of the basic size can be selected according to the actual situation.

[0140] When the second probability of the pixel point is less than the second preset threshold, it is considered that the first pixel point in the pathological section staining image i Row, No. j The pixels in the column tend to be pixels in the colon biopsy tissue area. In order to remove noise and retain the details of the colon biopsy tissue in the image as much as possible, the size of the structural element is set to 1, that is, the pixel is not opened.

[0141] It should also be noted that when performing the opening operation, if the size of the structural element used in the erosion process is the same as that used in the dilation process, the dilation process may restore the low-quality blurred area eroded by the erosion process. Therefore, in this example, the size of the structural element used for dilation should not be smaller than the size of the structural element used in the erosion process.

[0142] The following is a description of a device for identifying villus on the surface of colon mucosa in inflammatory bowel disease provided by the present invention. The device for identifying villus on the surface of colon mucosa in inflammatory bowel disease described below and the method for identifying villus on the surface of colon mucosa in inflammatory bowel disease described above can refer to each other.

[0143] like Figure 3 As shown, a device for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease includes a first determination module 301, a second determination module 302, an adjustment module 303 and an identification module 304:

[0144] A first determination module 301 is used to determine a first possibility that each pixel point in the pathological section staining image is located in the background area based on the grayscale image of the pathological section staining image of the colon biopsy tissue;

[0145] First, the pathological sections of the colon biopsy tissue are stained by H&E staining, and images of the pathological sections are captured using a microscope camera that can provide high-resolution images to obtain H&E staining images of the pathological sections of the colon biopsy tissue, which are collectively referred to as pathological section staining images in the following text.

[0146] It is understandable that the background area represents the location where the colon biopsy tissue does not appear in the pathological section staining image. In the grayscale image of the pathological section staining image, the background area is usually brighter than the area where the colon biopsy tissue appears, so in the grayscale image, the pixel value of the pixel point located in the background area is usually higher.

[0147] At the same time, the limitations of the optical imaging system lead to low-quality blurred areas such as Figure 2 The portion framed by the dotted line is generally a filamentous color block free around the colon biopsy tissue, whose gray value is lower than that of the background area, but lacks texture details.

[0148] On this basis, in order to reduce the impact of low-quality blurred areas in the pathological section staining image, it is necessary to make a preliminary judgment on whether each position in the image is in the staining area based on the grayscale value of the pixel point at each position in the image, and determine the first possibility that each pixel point in the pathological section staining image is located in the background area.

[0149] The higher the first possibility, the more likely the corresponding pixel point is to be the pixel point corresponding to the background area, and the smaller the first possibility, the more likely it is to be the pixel point corresponding to the biopsy tissue.

[0150] A second determination module 302 is used to calculate a second possibility that each pixel point is located in a low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image;

[0151] Furthermore, it can be understood that the biopsy tissue area in the pathological section staining image has richer texture information than the low-quality blurred area, that is, the different texture features of the colon biopsy tissue area and the low-quality blurred area in the grayscale image. Therefore, based on the pixel points that are more likely to be non-background areas determined based on the first possibility, it is possible to further identify whether the pixel points belong to the colon biopsy tissue area or the low-quality blurred area by judging their texture information.

[0152] Optionally, an edge image of the pathological section staining image is obtained by edge detection technology, and the richness of texture information in the area around each pixel is determined based on the edge image, so as to determine the second possibility that each pixel in the pathological section staining image is located in a low-quality blurred area based on the first possibility.

[0153] The greater the second possibility is, the greater the possibility that the pixel is located in a low-quality blurred area.

[0154] An adjustment module 303, configured to adjust the size of the structural element in the process of performing an operation on the pathological section staining image based on the second possibility;

[0155] It can be understood that by calculating the second possibility of each pixel in the pathological section staining image, that is, by screening the grayscale value of each pixel and the richness of the surrounding texture, pixels with lower grayscale values ​​and richer surrounding textures are obtained, and they are confirmed as pixels corresponding to the colon biopsy tissue area, and other pixels are confirmed as pixels corresponding to the background area and low-quality blurred areas.

[0156] On this basis, in the process of opening the pathological section staining image, for the pixel points corresponding to the biopsy tissue area, their structural elements are not processed; for the pixel points corresponding to the background area and the low-quality blurred area, their structural elements are enlarged for opening operation so as to remove them.

[0157] The identification module 304 is used to identify the degree of villi on the surface of the colon mucosa according to the size-adjusted pathological section staining image.

[0158] Each channel of the pathological section staining image is processed, and the processed channels are recombined into a color image to obtain the colon biopsy tissue area in the pathological section staining image and identify its villiformization degree.

[0159] At this time, the pixels corresponding to the background area and the low-quality blurred area are removed from the pathological section staining image after the opening operation, which to a certain extent avoids the process of identifying the degree of villus formation on the colon mucosal surface from being affected by the low-quality blurred area, thereby obtaining a more accurate recognition result of the degree of villus formation on the colon mucosal surface.

[0160] The present invention processes the pathological section staining image of the colon biopsy tissue, so that the structure element can have a larger size when it is located in the low-quality blurred area, so as to effectively remove the low-quality blurred area in the image, and can have a smaller size when the structure element is located in the area of ​​the colon biopsy tissue in the image. While effectively removing the low-quality blurred area in the staining image, the image details of the colon biopsy tissue in the staining image can also be retained, which is beneficial to the feature extraction of the colon biopsy tissue in the image, thereby improving the recognition accuracy of the villi on the surface of the colon mucosa in inflammatory bowel disease.

[0161] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430 and a communication bus 440, wherein the processor 410, the communication interface 420 and the memory 430 communicate with each other through the communication bus 440. The processor 410 may call the logic instructions in the memory 430 to execute a method for identifying the villus on the surface of the colon mucosa of inflammatory bowel disease, the method comprising: based on the grayscale image of the pathological section staining image of the colon biopsy tissue, determining the first possibility that each pixel point in the pathological section staining image is located in the background area; based on the first possibility and the texture feature of the grayscale image, calculating the second possibility that each pixel point is located in the low-quality fuzzy area of ​​the pathological section staining image; based on the second possibility, adjusting the size of the structural element in the process of operating the pathological section staining image; and identifying the degree of villus on the surface of the colon mucosa according to the pathological section staining image after the size adjustment.

[0162] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0163] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the method for identifying the villus on the surface of the colon mucosa in inflammatory bowel disease provided by the above-mentioned methods, the method including: based on the grayscale image of the pathological section staining image of the colon biopsy tissue, determining a first possibility that each pixel point in the pathological section staining image is located in the background area; based on the first possibility and the texture features of the grayscale image, calculating a second possibility that each pixel point is located in a low-quality fuzzy area of ​​the pathological section staining image; based on the second possibility, adjusting the size of the structural element during the operation of the pathological section staining image; and identifying the degree of villus on the surface of the colon mucosa according to the pathological section staining image after the size adjustment.

[0164] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the method for identifying the villus on the surface of the colon mucosa in inflammatory bowel disease provided by the above-mentioned methods, the method comprising: determining a first possibility that each pixel point in the pathological section staining image is located in a background area based on a grayscale image of a pathological section staining image of a colon biopsy tissue; calculating a second possibility that each pixel point is located in a low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and texture features of the grayscale image; adjusting the size of the structural element during the operation of the pathological section staining image based on the second possibility; and identifying the degree of villus on the surface of the colon mucosa according to the pathological section staining image after the size adjustment.

[0165] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0166] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0167] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying villi on the surface of colon mucosa in inflammatory bowel disease, characterized in that: include: Based on the grayscale image of the pathological section staining image of the colon biopsy tissue, determining a first possibility that each pixel point in the pathological section staining image is located in the background area; Calculating a second possibility that each pixel point is located in a low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image; adjusting the size of the structural element during the operation of the pathological section staining image based on the second possibility; The degree of villi on the surface of the colon mucosa is identified according to the size-adjusted staining image of the pathological section; The step of adjusting the size of the structural element in the process of performing an operation on the pathological section staining image based on the second possibility specifically includes: When the second possibility of the pixel point is not less than a second preset threshold, the size of the structure element corresponding to the pixel point is increased.

2. A method for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease according to claim 1, characterized in that: The step of determining the first possibility that each pixel point in the pathological section staining image is located in the background area based on the grayscale image of the pathological section staining image of the colon biopsy tissue specifically includes: Calculate the first pixel mean and the second pixel mean of the row and the column to which each pixel point in the grayscale image belongs respectively; The larger one of the first pixel mean and the second pixel mean of each pixel in the grayscale image after normalization is used as the first possibility of each pixel in the pathological section staining image.

3. A method for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease according to claim 1, characterized in that: The step of determining the first possibility that each pixel point in the pathological section staining image is located in the background area based on the grayscale image of the pathological section staining image of the colon biopsy tissue specifically includes: Calculate the first pixel mean and the second pixel mean of the row and the column to which each pixel point in the grayscale image belongs respectively; The larger one of the first pixel mean and the second pixel mean of each pixel in the grayscale image is normalized as the third possibility of each pixel in the grayscale image; Calculate the ratio of the pixel value of each pixel in the grayscale image to 255 as the first weight; The product of the third possibility of each pixel in the grayscale image and the first weight is used as the first possibility of each pixel in the pathological section staining image.

4. A method for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease according to any one of claims 1 to 3, characterized in that: The step of calculating the second possibility that each pixel point is located in the low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image specifically includes: Acquire an edge image of a staining image of a pathological section of the colon biopsy tissue; For each pixel point in the edge image, determining a first variance of pixels within a preset range centered on each pixel point; Calculating the second variance of all pixels in the edge image; The second possibility of each pixel point in the pathological section staining image is determined according to the ratio of the first variance to the second variance and the first possibility.

5. A method for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease according to claim 4, characterized in that: The step of calculating the second possibility that each pixel point is located in the low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image also includes: For each pixel point in the pathological section staining image, when the first possibility of the pixel point is not less than a first preset threshold, the second possibility of the pixel point is increased.

6. A method for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease according to claim 4, characterized in that: The step of calculating the second possibility that each pixel point is located in the low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image also includes: Calculate a first average value of the number of edge pixels within all the preset ranges; For each pixel point in the pathological section staining image, when the number of edge pixels within the preset range centered at the pixel point is less than the first mean value, the second possibility of the pixel point is increased.

7. A device for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease, characterized in that: include: A first determination module is used to determine a first possibility that each pixel point in the pathological section staining image is located in the background area based on the grayscale image of the pathological section staining image of the colon biopsy tissue; A second determination module is used to calculate a second possibility that each pixel point is located in a low-quality fuzzy area of ​​the pathological section staining image based on the first possibility and the texture feature of the grayscale image; An adjustment module, configured to adjust the size of the structural element in the process of operating the pathological section staining image based on the second possibility; An identification module, used to identify the degree of villi on the surface of the colon mucosa according to the size-adjusted pathological section staining image; The adjustment module is specifically used for: When the second possibility of the pixel point is not less than a second preset threshold, the size of the structure element corresponding to the pixel point is increased.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, a method for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease as described in any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for identifying villus formation on the surface of colon mucosa in inflammatory bowel disease as claimed in any one of claims 1 to 6 is implemented.

Images