Physiological tissue lesion recognition method, device and equipment and readable storage medium

Abstract

The invention discloses a physiological tissue lesion recognition method, device and equipment and a readable storage medium. The method comprises the steps that a surface area image of a physiological tissue is acquired; analyzing the surface area image, and determining a surface feature corresponding to the physiological tissue; and based on the surface features and the type of the physiological tissue, determining a target lesion corresponding to the physiological tissue. By combining the surface features of the surface area image of the physiological tissue and the type of the physiological tissue, the pathological change of the physiological tissue is determined, and the accuracy of identifying the pathological change of the physiological tissue based on the surface features can be improved.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to a method, device, equipment and readable storage medium for identifying physiological tissue lesions. Background Art

[0002] With the rapid development of AI technology, AI-assisted diagnosis technology has gradually been applied to the screening and diagnosis of medical lesions. After scanning the human body with a scanning device, AI-assisted diagnosis technology can identify some lesion-related information in the medical image as a reference for doctors' diagnosis.

[0003] However, during the detection of physiological tissues and organs, since there may be uneven areas on the surface of the physiological tissues, these uneven areas may be normal physiological structures or may contain lesions. It is impossible to identify whether the physiological tissues have lesions based on the surface characteristics of the physiological tissues.

[0004] Therefore, how to improve the accuracy of identifying physiological tissue lesions based on surface features is an urgent problem to be solved. Summary of the invention

[0005] The embodiments of the present application provide a method, device, equipment and readable storage medium for identifying physiological tissue lesions, which can improve the accuracy of identifying physiological tissue lesions based on surface features.

[0006] In a first aspect, an embodiment of the present application provides a method for identifying physiological tissue lesions, the method comprising:

[0007] Acquire surface area images of physiological tissues;

[0008] Analyzing the surface area image to determine surface features corresponding to the physiological tissue;

[0009] Based on the surface features and the type of the physiological tissue, a target lesion corresponding to the physiological tissue is determined.

[0010] Optionally, analyzing the surface area image to determine the surface features corresponding to the physiological tissue includes:

[0011] Performing smoothing processing on the surface area in the surface area image to obtain a smooth surface area;

[0012] The surface area is compared with the smooth surface area to determine the corresponding surface features of the physiological tissue; the surface features include the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area.

[0013] Optionally, analyzing the surface area image to determine the surface features corresponding to the physiological tissue includes:

[0014] For each pixel point in the surface area image, obtain a first coordinate of the pixel point and a second coordinate of a pixel point adjacent to the pixel point, and determine a direction field of the pixel point according to the first coordinate and the second coordinate;

[0015] Based on the direction field corresponding to each pixel point in the surface area image, the corresponding surface features of the physiological tissue are determined; the surface features include the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area.

[0016] Optionally, determining a target lesion corresponding to the physiological tissue based on the surface feature and the type of the physiological tissue includes:

[0017] Based on the type of the physiological tissue, query a preset physiological tissue lesion surface feature database to obtain the lesion surface feature corresponding to the type of the physiological tissue;

[0018] Based on the lesion surface feature and the surface feature, a target lesion corresponding to the physiological tissue is determined.

[0019] Optionally, determining the target lesion corresponding to the physiological tissue based on the lesion surface feature and the surface feature includes:

[0020] The roughness of the surface area in the surface feature, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area are counted and quantified to determine a first evaluation index corresponding to the surface feature;

[0021] The unevenness of the lesion surface area, the distribution density of the non-smooth area of ​​the lesion surface area, and the area of ​​the non-smooth area in each of the lesion surface features are counted and quantified to determine a second evaluation index corresponding to the lesion surface feature;

[0022] The similarity between the first evaluation indicator and each of the second evaluation indicators is calculated respectively, a target evaluation indicator is determined based on the similarity, and the lesion corresponding to the target evaluation indicator is determined as the target lesion corresponding to the physiological tissue, and the target evaluation indicator is the second evaluation indicator having the maximum similarity with the first evaluation indicator and the similarity is greater than a preset similarity threshold.

[0023] Optionally, acquiring a surface area image of a physiological tissue comprises:

[0024] Acquire the type and image of physiological tissue;

[0025] Selecting a target segmentation model from a preset segmentation model library based on the type of the physiological tissue, and segmenting the physiological tissue image based on the target segmentation model to obtain a physiological tissue segmentation image;

[0026] A surface area image of the physiological tissue is extracted from the physiological tissue segmentation image.

[0027] Optionally, before analyzing the surface area image to determine the surface features corresponding to the physiological tissue, the method further comprises:

[0028] Constructing a buffer zone in the surface area image, and calculating edge features of each pixel point in the buffer zone;

[0029] The clarity of the surface area boundary of the surface area image is optimized based on the edge feature to obtain an optimized surface area image.

[0030] In a second aspect, the present application provides a physiological tissue lesion identification device.

[0031] An acquisition module, used for acquiring a surface area image of a physiological tissue;

[0032] An analysis module, used for analyzing the surface area image to determine the surface features corresponding to the physiological tissue;

[0033] A determination module is used to determine a target lesion corresponding to the physiological tissue based on the surface feature and the type of the physiological tissue.

[0034] In a third aspect, an embodiment of the present application also provides a physiological tissue lesion identification device, comprising a memory storing a computer program; a processor loads the computer program from the memory to execute the steps of any physiological tissue lesion identification method provided in the embodiment of the present application.

[0035] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores a computer program, and the computer program is suitable for loading by a processor to execute the steps of any physiological tissue lesion identification method provided in the embodiment of the present application.

[0036] In a fifth aspect, an embodiment of the present application further provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of any one of the physiological tissue lesion identification methods provided in the embodiments of the present application.

[0037] By adopting the scheme of the embodiment of the application, a surface area image of a physiological tissue is acquired; the surface area image is analyzed to determine the surface features corresponding to the physiological tissue; and based on the surface features and the type of the physiological tissue, a target lesion corresponding to the physiological tissue is determined. By combining the surface features of the surface area image of the physiological tissue and the type of the physiological tissue to determine the lesion of the physiological tissue, the accuracy of identifying the lesion of the physiological tissue based on the surface features can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0039] Figure 1 This is a schematic flow chart of a first embodiment of a method for identifying physiological tissue lesions provided by the present application;

[0040] Figure 2 This is a schematic flow chart of a second embodiment of the physiological tissue lesion identification method provided by the present application;

[0041] Figure 3 It is a schematic flow chart of a third embodiment of the physiological tissue lesion identification method provided by the present application;

[0042] Figure 4 It is a schematic flow chart of a fourth embodiment of the physiological tissue lesion identification method provided by the present application;

[0043] Figure 5 It is a schematic diagram of the surface characteristics of liver lesions provided by the present application;

[0044] Figure 6 It is a schematic flow chart of a fifth embodiment of the physiological tissue lesion identification method provided by the present application;

[0045] Figure 7 is a schematic diagram of the structure of a physiological tissue lesion identification device provided in an embodiment of the present application;

[0046] Figure 8 It is a structural schematic diagram of the physiological tissue lesion identification device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application. At the same time, in the description of the embodiments of the present application, the terms "first", "second", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more features. In the description of the embodiments of the present application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0048] Embodiments of the present application provide a method, apparatus, device and readable storage medium for identifying physiological tissue lesions.

[0049] Specifically, this embodiment will be described from the perspective of a physiological tissue lesion identification device, which can be specifically integrated into a physiological tissue lesion identification apparatus, that is, the physiological tissue lesion identification method of the embodiment of the present application can be executed by the physiological tissue lesion identification device.

[0050] The physiological tissue lesion identification method provided in the embodiment of the present application can be applied to a physiological tissue lesion identification device, which can be a smart terminal, a PC terminal, a mobile terminal, or other device.

[0051] The following is a detailed description in conjunction with the accompanying drawings. In this embodiment, the execution subject is taken as an example of a physiological tissue lesion identification device. It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments. Although the logical order is shown in the flow chart, in some cases, the steps shown or described may be performed in an order different from that shown in the accompanying drawings.

[0052] Please refer to Figure 1 , a first embodiment of the physiological tissue lesion identification method is proposed, and the first embodiment includes the following steps:

[0053] Step 101, acquiring a surface area image of a physiological tissue;

[0054] Step 102, analyzing the surface area image to determine the surface features corresponding to the physiological tissue;

[0055] Step 103: determining a target lesion corresponding to the physiological tissue based on the surface feature and the type of the physiological tissue.

[0056] In this embodiment, when it is necessary to identify the lesion of a certain physiological tissue, the relevant user inputs the medical image of the physiological tissue into the physiological tissue lesion identification device; after the physiological tissue lesion identification device obtains the medical image of the physiological tissue, it extracts the surface area image corresponding to the medical image of the physiological tissue; the physiological tissue lesion identification device performs feature analysis on the surface area image to determine the surface features corresponding to the physiological tissue; the physiological tissue lesion identification device determines the target lesion corresponding to the physiological tissue based on the surface features and the type of physiological tissue.

[0057] It should be noted that the physiological tissue may be the heart, liver, kidney, etc.; the medical image of the physiological tissue and the surface area image of the physiological tissue may be a two-dimensional medical image or a three-dimensional medical image, preferably a three-dimensional medical image, based on which more accurate surface features corresponding to the physiological tissue may be obtained.

[0058] The physiological tissue lesion recognition device of this embodiment acquires a surface area image of the physiological tissue; analyzes the surface area image to determine the surface features corresponding to the physiological tissue; and determines the target lesion corresponding to the physiological tissue based on the surface features and the type of the physiological tissue. By combining the surface features of the surface area image of the physiological tissue and the type of the physiological tissue to determine the lesion of the physiological tissue, the accuracy of identifying the lesion of the physiological tissue based on the surface features can be improved.

[0059] Specifically, each step is described in detail below:

[0060] Step 101, acquiring a surface area image of a physiological tissue;

[0061] In this step, the physiological tissue lesion recognition device has an interactive module for interacting with the user, and the user can input the physiological tissue medical image that needs to be lesion recognized into the physiological tissue lesion recognition device based on the interactive module, and the physiological tissue lesion recognition device obtains the physiological tissue medical image and extracts the surface area image of the physiological tissue in the physiological tissue medical image. It can be understood that the surface area image of the physiological tissue only includes the surface contour in the physiological tissue medical image.

[0062] Specifically, step 101 includes:

[0063] Step 1011, obtaining the type of physiological tissue and the physiological tissue image;

[0064] In this step, the physiological tissue lesion recognition device has an interactive module for interacting with the user, and the user can input the type of physiological tissue and the physiological tissue image into the physiological tissue lesion recognition device based on the interactive module. Optionally, the physiological tissue lesion recognition device can identify and determine the type of corresponding physiological tissue based on the physiological tissue image input by the user.

[0065] Step 1012, selecting a target segmentation model from a preset segmentation model library based on the type of the physiological tissue, and segmenting the physiological tissue image based on the target segmentation model to obtain a physiological tissue segmentation image;

[0066] In this step, after determining the type of physiological tissue, the physiological tissue lesion recognition device selects a target segmentation model from a preset segmentation model library based on the type of physiological tissue, and segments the physiological tissue image based on the target segmentation model to obtain a physiological tissue segmentation image. It is understandable that due to the different shapes and sizes of different physiological tissues, their corresponding segmentation models will also be different. In order to improve the accuracy of physiological tissue segmentation, a segmentation model for each physiological tissue is created in advance and stored in a preset segmentation model library in the physiological tissue lesion recognition device. When the physiological tissue lesion recognition device determines the type of physiological tissue, it can select a corresponding target segmentation model from the preset segmentation model library according to the type of physiological tissue for segmenting the physiological tissue image to obtain a physiological tissue segmentation image.

[0067] Step 1013: extracting the surface area image of the physiological tissue in the physiological tissue segmentation image.

[0068] In this step, the physiological tissue lesion recognition device extracts the surface area image of the physiological tissue in the physiological tissue segmentation image. Optionally, the physiological tissue image is usually a three-dimensional medical image, and the physiological tissue segmentation image obtained by segmenting the physiological tissue image with the segmentation model includes a surface area image and an internal area image. The physiological tissue lesion recognition device can directly obtain the surface area image of the physiological tissue. Optionally, the physiological tissue segmentation image obtained by segmenting the physiological tissue image with the segmentation model includes a plurality of physiological tissue partial images, and the surface area extraction is performed on each physiological tissue partial image to form a surface area image of the physiological tissue. By segmenting into a plurality of physiological tissue partial images, more details of the surface area of ​​the physiological tissue can be further obtained, thereby improving the accuracy of the extracted surface area image of the physiological tissue.

[0069] Step 102, analyzing the surface area image to determine the surface features corresponding to the physiological tissue;

[0070] In this step, after extracting the surface area image of the physiological tissue, the physiological tissue lesion recognition device analyzes the surface area image to determine the surface features corresponding to the physiological tissue. Optionally, the physiological tissue lesion recognition device uses an edge smoothing algorithm to smooth the surface area image to obtain a smooth surface area, and determines the surface features corresponding to the physiological tissue based on the comparison between the smooth surface area and the surface area image. Optionally, the physiological tissue lesion recognition device creates a coordinate system on the surface area image, extracts the coordinates of each pixel point on the surface area image, determines the direction field of each pixel point according to the coordinates of each pixel point, and determines the surface features corresponding to the physiological tissue based on the direction field. Optionally, the physiological tissue lesion recognition device determines the gradient field of each pixel point according to the coordinates of each pixel point, and determines the surface features corresponding to the physiological tissue based on the gradient field. Among them, the direction field describes the direction of the vector at each pixel, but does not necessarily describe the magnitude of the vector. In a direction field, the length of the vector can be different, but at any given point, all vectors point in the same direction. The gradient field is a special type of vector field, in which the magnitude and direction of the vector are determined by the gradient of a scalar function. In other words, the gradient field describes the rate of change and direction of a scalar field at each pixel point.

[0071] Step 103: determining a target lesion corresponding to the physiological tissue based on the surface feature and the type of the physiological tissue.

[0072] In this step, the physiological tissue lesion recognition device determines the target lesion corresponding to the physiological tissue based on the surface features and the type of the physiological tissue. Specifically, the physiological tissue lesion recognition device queries a preset physiological tissue lesion surface feature database based on the type of the physiological tissue, obtains the lesion surface features corresponding to the type of the physiological tissue, and determines the target lesion corresponding to the physiological tissue based on the lesion surface features and the surface features.

[0073] The physiological tissue lesion recognition device of this embodiment acquires a surface area image of the physiological tissue; analyzes the surface area image to determine the surface features corresponding to the physiological tissue; and determines the target lesion corresponding to the physiological tissue based on the surface features and the type of the physiological tissue. By combining the surface features of the surface area image of the physiological tissue and the type of the physiological tissue to determine the lesion of the physiological tissue, the accuracy of identifying the lesion of the physiological tissue based on the surface features can be improved.

[0074] Further, refer to Figure 2 A second embodiment of the physiological tissue lesion identification method is proposed. The difference between the second embodiment and the first embodiment is that the surface area image is analyzed to determine the surface features corresponding to the physiological tissue, including:

[0075] Step 1021, performing smoothing processing on the surface area in the surface area image to obtain a smoothed surface area;

[0076] In this step, the physiological tissue lesion recognition device smoothes the surface area in the surface area image to obtain a smooth surface area. Specifically, the physiological tissue lesion recognition device uses an edge detection algorithm (such as Sobel, Canny, etc.) to extract all pixel points on the surface area in the surface area image, connects the detected pixel points into continuous edge lines according to a preset connection method, and smoothes all edge lines to reduce the jaggedness and discontinuity of the edge lines. The specific edge line smoothing algorithm includes: Gaussian filtering: Gaussian blurring the edge line, and eliminating noise and small edge discontinuities by weighted averaging the grayscale values ​​of nearby pixels. Median filtering: Replacing the value of the edge point with the median of the pixel values ​​around the edge point can effectively remove isolated noise points. Bilateral filtering: Combining spatial distance and grayscale value similarity to smooth the edge, retaining the clarity of the edge while reducing noise. Average filtering: Simply taking the average value of the surrounding pixels to replace the value of the edge point, suitable for situations with less noise. The physiological tissue lesion recognition device may adopt one or more of the above algorithms to smooth each edge line, and merge all the smoothed edge lines to obtain a smooth surface area.

[0077] Step 1022, compare the surface area with the smooth surface area to determine the corresponding surface features of the physiological tissue; the surface features include the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area.

[0078] In this step, the physiological tissue lesion recognition device compares the surface area with the smooth surface area to determine the corresponding surface features of the physiological tissue; wherein the surface features include the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area. Specifically, the physiological tissue lesion recognition device takes the smooth surface area as a reference, compares each pixel point in the smooth surface area with each pixel point in the surface area, and calculates the degree of fluctuation of each pixel point in the surface area relative to the corresponding pixel point in the smooth surface area, and the degree of fluctuation represents the height difference of each pixel point in the surface area relative to the corresponding pixel point in the smooth surface area; the physiological tissue lesion recognition device calculates the roughness of the surface area based on the degree of fluctuation of each pixel point in the surface area, and at the same time, determines the area composed of pixels in the surface area whose degree of fluctuation is greater than the preset degree of fluctuation as a non-smooth area, and statistically determines the distribution density of the non-smooth area of ​​the surface area and the area of ​​the non-smooth area.

[0079] The physiological tissue lesion recognition device in this embodiment performs smoothing processing on the surface area in the surface area image to obtain a smooth surface area, and compares the surface area with the smooth surface area to determine the corresponding surface features of the physiological tissue. This can improve the accuracy of determining the corresponding surface features of the physiological tissue, which helps to improve the accuracy of subsequent recognition of physiological tissue lesions based on surface features.

[0080] Further, refer to Figure 3 A third embodiment of the physiological tissue lesion identification method is proposed. The difference between the third embodiment and the first embodiment and the second embodiment is that the surface area image is analyzed to determine the surface features corresponding to the physiological tissue, including:

[0081] Step 1023, for each pixel in the surface area image, obtain a first coordinate of the pixel and a second coordinate of a pixel adjacent to the pixel, and determine a direction field of the pixel according to the first coordinate and the second coordinate;

[0082] In this step, the physiological tissue lesion recognition device establishes a coordinate system on the surface area image, obtains the first coordinate of the pixel point and the second coordinate of the pixel point's adjacent pixel point for each pixel point in the surface area image, and determines the direction field of the pixel point based on the first coordinate and the second coordinate. The direction field of the pixel point represents the degree of change in the direction of the pixel point relative to the adjacent pixel points.

[0083] Specifically, the physiological tissue lesion recognition device takes one of the pixels as the target pixel, obtains all the adjacent pixels to the target pixel, and sequentially groups each adjacent pixel with the target pixel into a group, and calculates the reference direction vector of the target pixel according to the second coordinates of the adjacent pixels in the group and the first coordinates of the target pixel, until all the reference direction vectors corresponding to the target pixel are calculated based on the second coordinates of each adjacent pixel and the first coordinates of the target pixel, and the physiological tissue lesion recognition device calculates the mean of all the reference direction vectors corresponding to the target pixel to determine the target direction vector corresponding to the target pixel. Based on the above method, the physiological tissue lesion recognition device can determine the target direction vector corresponding to each pixel on the surface area image. For each pixel on the surface area image, the physiological tissue lesion recognition device inputs the target direction vector of each pixel and the preset direction vector into the calculation model for calculating the direction field, and the calculation model calculates the direction field corresponding to each pixel based on the target direction vector and the preset direction vector of each pixel.

[0084] Step 1024, based on the direction field corresponding to each pixel point in the surface area image, determine the corresponding surface features of the physiological tissue; the surface features include the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area.

[0085] In this step, the physiological tissue lesion identification device determines the corresponding surface features of the physiological tissue based on the direction field corresponding to each pixel point in the surface area image, wherein the surface features include the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area.

[0086] Specifically, the direction field of a pixel point represents the degree of direction change of the pixel point relative to adjacent pixel points. The physiological tissue lesion identification device calculates the roughness of the surface area based on the degree of direction change of each pixel point in the surface area. At the same time, the area composed of pixels in the surface area whose degree of direction change is greater than a preset degree of direction change is determined as a non-smooth area, and the distribution density of the non-smooth area of ​​the surface area and the area of ​​the non-smooth area are statistically determined.

[0087] The physiological tissue lesion recognition device in this embodiment obtains the first coordinate of each pixel point in the surface area image and the second coordinate of the pixel point adjacent to the pixel point, and determines the direction field of the pixel point based on the first coordinate and the second coordinate; based on the direction field corresponding to each pixel point in the surface area image, the corresponding surface feature of the physiological tissue is determined. The accuracy of determining the corresponding surface feature of the physiological tissue can be improved, which helps to improve the accuracy of subsequent recognition of physiological tissue lesions based on the surface feature.

[0088] Further, refer to Figure 4 , a fourth embodiment of the physiological tissue lesion identification method is proposed. The difference between the fourth embodiment and the first to third embodiments is that the target lesion corresponding to the physiological tissue is determined based on the surface feature and the type of the physiological tissue, including:

[0089] Step 1031, based on the type of the physiological tissue, query a preset physiological tissue lesion surface feature database to obtain the lesion surface feature corresponding to the type of the physiological tissue;

[0090] In this step, after determining the surface features corresponding to the physiological tissue and the type of the physiological tissue, the physiological tissue lesion recognition device queries the preset physiological tissue lesion surface feature database based on the type of the physiological tissue to obtain the lesion surface features corresponding to the type of the physiological tissue. It can be understood that the lesion surface features corresponding to the occurrence of lesions of different types of physiological tissues are different. The physiological tissue lesion recognition device stores the lesion surface features corresponding to the occurrence of lesions of different types of physiological tissues in advance in the preset physiological tissue lesion surface feature database, such as: lesion surface features of the liver in cirrhosis, lesion surface features of the liver in ascites, lesion surface features of the heart in myocarditis, lesion surface features of the kidney in uremia, etc. The physiological tissue lesion recognition device compares the type of the physiological tissue with all the types of physiological tissues in the preset physiological tissue lesion surface feature database to obtain the lesion surface features corresponding to the type of the physiological tissue. For example, if the current type of the physiological tissue is liver, the physiological tissue lesion recognition device can obtain the lesion surface features corresponding to the liver from the preset physiological tissue lesion surface feature database, including: lesion surface features of the liver in cirrhosis, lesion surface features of the liver in ascites, etc.

[0091] Step 1032: Determine a target lesion corresponding to the physiological tissue based on the lesion surface feature and the surface feature.

[0092] In this step, after obtaining the lesion surface features corresponding to the type of physiological tissue, the physiological tissue lesion recognition device determines the target lesion corresponding to the physiological tissue based on the lesion surface features and the surface features. It can be understood that the lesion surface features corresponding to the type of physiological tissue include surface features corresponding to multiple lesions. The lesion surface features corresponding to each lesion are compared with the surface features to determine the target lesion corresponding to the physiological tissue.

[0093] Specifically, step 1032 includes:

[0094] Step 10321, counting and quantifying the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area in the surface feature to determine a first evaluation index corresponding to the surface feature;

[0095] In this step, the physiological tissue lesion identification device counts and quantifies the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area in the surface features, and determines the first evaluation index corresponding to the surface features. The first evaluation index includes a first roughness index, a first density index, and a first area index.

[0096] Specifically, the physiological tissue lesion identification device determines a first roughness index by mathematically modeling and analyzing the roughness of the surface area. The first roughness index can be represented by the average curvature of the surface, the fractal dimension of the surface, or the convexity and concavity of the surface; the physiological tissue lesion identification device counts the total area of ​​the surface area and the number of non-smooth areas, and then compares the number of non-smooth areas with the total area of ​​the surface area to determine the first density index; the physiological tissue lesion identification device calculates the area of ​​each non-smooth area in the surface area, adds the area of ​​each non-smooth area, and obtains the first area index.

[0097] Step 10322, respectively counting and quantifying the unevenness of the lesion surface area, the distribution density of the non-smooth area of ​​the lesion surface area, and the area of ​​the non-smooth area in each of the lesion surface features, to determine a second evaluation index corresponding to the lesion surface feature;

[0098] In this step, the physiological tissue lesion identification device counts and quantifies the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area in each lesion surface feature, and determines the second evaluation index corresponding to the surface feature. The second evaluation index includes a second roughness index, a second density index, and a second area index.

[0099] Specifically, for each lesion surface feature, the physiological tissue lesion identification device determines a second roughness index by mathematically modeling and analyzing the roughness of the lesion surface area. The second roughness index can be represented by the average curvature of the surface, the fractal dimension of the surface, or the convexity and concavity of the surface; the physiological tissue lesion identification device counts the total area of ​​the lesion surface area and the number of non-smooth areas, and then compares the number of non-smooth areas with the total area of ​​the lesion surface area to determine the second density index; the physiological tissue lesion identification device calculates the area of ​​each non-smooth area in the lesion surface area, adds the area of ​​each non-smooth area, and obtains the second area index.

[0100] Step 10323, respectively calculate the similarity between the first evaluation indicator and each of the second evaluation indicators, determine the target evaluation indicator based on the similarity, and determine the lesion corresponding to the target evaluation indicator as the target lesion corresponding to the physiological tissue, and the target evaluation indicator is the second evaluation indicator with the maximum similarity with the first evaluation indicator and the similarity is greater than a preset similarity threshold.

[0101] In this step, the physiological tissue lesion identification device respectively calculates the similarity between the first evaluation indicator and each second evaluation indicator, determines the target evaluation indicator based on the similarity, and determines the lesion corresponding to the target evaluation indicator as the target lesion corresponding to the physiological tissue. The target evaluation indicator is the second evaluation indicator that has the greatest similarity with the first evaluation indicator and the similarity is greater than a preset similarity threshold.

[0102] Specifically, for each lesion surface feature, the physiological tissue lesion identification device calculates a first similarity between a first roughness index and a corresponding second roughness index, calculates a second similarity between a first density index and a corresponding second density index, calculates a third similarity between a first area index and a corresponding second area index, and compares the first similarity, the second similarity and the third similarity with a preset similarity threshold value, respectively. If it is determined that the first similarity, the second similarity and the third similarity are all greater than the preset similarity threshold value, then the second roughness index, the second density index and the second area index corresponding to the current lesion surface feature can be determined as target evaluation indicators, and then the lesion corresponding to the target evaluation indicator is determined to be the target lesion corresponding to the physiological tissue.

[0103] For example, Figure 5 As shown, different lesions correspond to different liver surface features. The lesions may be nodules that are perfectly round and evenly distributed over the entire liver surface. Nodules of different degrees of lesions may be of different sizes. Lesions may also be nodules of different sizes and diaphragms of different thicknesses that exist in the same or different tissue regions on the liver surface. For example, in case of liver fibrosis, the surface characteristics of the liver in the first stage of liver fibrosis are shown in the upper right figure, where the nodules are perfectly round and evenly distributed on the entire liver surface, and the lesion nodules are relatively small. In case of liver cirrhosis, the surface characteristics of the liver in case of liver cirrhosis are shown in the lower right figure, where nodules of different sizes and diaphragms of different thicknesses exist in the same or different tissue regions on the liver surface. The physiological tissue lesion recognition device can determine the second roughness index, the second density index, and the second area index of each liver lesion according to the different surface characteristics of these liver lesions. At the same time, the physiological tissue lesion recognition device obtains the first roughness index, the first density index, and the first area index of the liver surface area image, and compares them with the second roughness index, the second density index, and the second area index of each liver lesion based on the first roughness index, the first density index, and the first area index, respectively, to finally determine whether the target lesion that may exist in the current liver surface area image is the first stage of liver fibrosis, the second stage of liver fibrosis, or the liver cirrhosis lesion.

[0104] The physiological tissue lesion recognition device of this embodiment queries a preset physiological tissue lesion surface feature database based on the type of physiological tissue to obtain the lesion surface feature corresponding to the type of physiological tissue; based on the lesion surface feature and the surface feature, the target lesion corresponding to the physiological tissue is determined. By comparing the lesion surface feature and the surface feature, the target lesion corresponding to the physiological tissue can be determined, thereby improving the accuracy of identifying physiological tissue lesions based on surface features.

[0105] Further, refer to Figure 6 , a fifth embodiment of the physiological tissue lesion identification method is proposed. The difference between the fifth embodiment and the first to fourth embodiments is that before analyzing the surface area image to determine the surface features corresponding to the physiological tissue, the method includes:

[0106] Step a, constructing a buffer zone in the surface area image, and calculating the edge feature of each pixel in the buffer zone;

[0107] Step b: optimizing the clarity of the surface area boundary of the surface area image based on the edge feature to obtain an optimized surface area image.

[0108] In step a to step b, the physiological tissue lesion recognition device analyzes the surface area image to determine the surface features corresponding to the physiological tissue. Due to reasons such as the accuracy of the segmentation model, the edge area of ​​the physiological tissue is blurred and there are false uneven areas (normal areas are segmented into concave and convex boundaries). At this time, in order to avoid the above problems, the physiological tissue lesion recognition device constructs a buffer zone in the surface area image, calculates the edge features of each pixel in the buffer zone, and optimizes the clarity of the surface area boundary of the surface area image based on the edge features to obtain an optimized surface area image.

[0109] Specifically, the physiological tissue lesion recognition device constructs a buffer with a preset width at the edge of the surface area image, and obtains the edge features of each pixel in the buffer through edge detection algorithms such as Sobel and Canny; in some cases, the extracted edge may not be a continuous line, and the physiological tissue lesion recognition device uses edge connection technology (such as Hough transform) to identify and connect disconnected edge segments to obtain more continuous and accurate edges. After determining the continuous and accurate edges, the physiological tissue lesion recognition device enhances the edge pixel intensity or applies a filter to improve the clarity and contrast of the edge features; the physiological tissue lesion recognition device then optimizes the edge clarity by increasing the sharpness of the pixels or further filtering. Finally, the physiological tissue lesion recognition device generates an optimized surface area image based on the optimized edge features.

[0110] The physiological tissue lesion recognition device of this embodiment constructs a buffer zone in the surface area image and calculates the edge features of each pixel in the buffer zone; based on the edge features, the clarity of the surface area boundary of the surface area image is optimized to obtain an optimized surface area image. This avoids the problem of false uneven areas caused by the blurring of the edge area of ​​the physiological tissue due to the accuracy of the segmentation model, etc., and can improve the accuracy of subsequent extraction of surface features, thereby improving the accuracy of identifying physiological tissue lesions based on surface features.

[0111] This embodiment also provides a physiological tissue lesion recognition device, which can be integrated into physiological tissue lesion recognition devices such as smart terminals, PC terminals, and mobile terminals. Figure 7 As shown, the physiological tissue lesion identification device may include:

[0112] An acquisition module 1001 is used to acquire a surface area image of a physiological tissue;

[0113] An analysis module 1002 is used to analyze the surface area image to determine the surface features corresponding to the physiological tissue;

[0114] The determination module 1003 is used to determine the target lesion corresponding to the physiological tissue based on the surface feature and the type of the physiological tissue.

[0115] In an optional example, the analysis module is further configured to:

[0116] Performing smoothing processing on the surface area in the surface area image to obtain a smooth surface area;

[0117] The surface area is compared with the smooth surface area to determine the corresponding surface features of the physiological tissue; the surface features include the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area.

[0118] In an optional example, the analysis module is further configured to:

[0119] For each pixel point in the surface area image, obtain a first coordinate of the pixel point and a second coordinate of a pixel point adjacent to the pixel point, and determine a direction field of the pixel point according to the first coordinate and the second coordinate;

[0120] Based on the direction field corresponding to each pixel point in the surface area image, the corresponding surface features of the physiological tissue are determined; the surface features include the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area.

[0121] In an optional example, the determining module is further configured to:

[0122] Based on the type of the physiological tissue, query a preset physiological tissue lesion surface feature database to obtain the lesion surface feature corresponding to the type of the physiological tissue;

[0123] Based on the lesion surface feature and the surface feature, a target lesion corresponding to the physiological tissue is determined.

[0124] In an optional example, the determining module is further configured to:

[0125] The roughness of the surface area in the surface feature, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area are counted and quantified to determine a first evaluation index corresponding to the surface feature;

[0126] The unevenness of the lesion surface area, the distribution density of the non-smooth area of ​​the lesion surface area, and the area of ​​the non-smooth area in each of the lesion surface features are counted and quantified to determine a second evaluation index corresponding to the lesion surface feature;

[0127] The similarity between the first evaluation indicator and each of the second evaluation indicators is calculated respectively, a target evaluation indicator is determined based on the similarity, and the lesion corresponding to the target evaluation indicator is determined as the target lesion corresponding to the physiological tissue, and the target evaluation indicator is the second evaluation indicator having the maximum similarity with the first evaluation indicator and the similarity is greater than a preset similarity threshold.

[0128] In an optional example, the acquisition module is also used to:

[0129] Acquire the type and image of physiological tissue;

[0130] Selecting a target segmentation model from a preset segmentation model library based on the type of the physiological tissue, and segmenting the physiological tissue image based on the target segmentation model to obtain a physiological tissue segmentation image;

[0131] A surface area image of the physiological tissue is extracted from the physiological tissue segmentation image.

[0132] In an optional example, the physiological tissue lesion identification device further includes an optimization module, and the optimization module is used to:

[0133] Constructing a buffer zone in the surface area image, and calculating edge features of each pixel point in the buffer zone;

[0134] The clarity of the surface area boundary of the surface area image is optimized based on the edge feature to obtain an optimized surface area image.

[0135] By adopting the solution of this embodiment, a surface area image of a physiological tissue is acquired; the surface area image is analyzed to determine the surface features corresponding to the physiological tissue; and based on the surface features and the type of the physiological tissue, a target lesion corresponding to the physiological tissue is determined. By combining the surface features of the surface area image of the physiological tissue and the type of the physiological tissue to determine the lesion of the physiological tissue, the accuracy of identifying the lesion of the physiological tissue based on the surface features can be improved.

[0136] Accordingly, the present application also provides a physiological tissue lesion identification device, such as Figure 8 As shown, Figure 8 A schematic diagram of the structure of a physiological tissue lesion identification device provided in an embodiment of the present application. The physiological tissue lesion identification device 1100 includes a processor 1101 having one or more processing cores, a memory 1102 having one or more computer-readable storage media, and a computer program stored in the memory 1102 and executable on the processor. The processor 1101 is electrically connected to the memory 1102. Those skilled in the art will appreciate that the structure of the physiological tissue lesion identification device shown in the figure does not constitute a limitation on the physiological tissue lesion identification device, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0137] The processor 1101 is the control center of the physiological tissue lesion identification device 1100. It uses various interfaces and lines to connect various parts of the entire physiological tissue lesion identification device 1100, and executes various functions and processes data of the physiological tissue lesion identification device 1100 by running or loading software programs and / or units stored in the memory 1102, and calling data stored in the memory 1102, so as to monitor the physiological tissue lesion identification device 1100 as a whole. The processor 1101 can be a processor CPU, a graphics processor GPU, a network processor (Network Processor, NP), etc., and can implement or execute the various methods, steps and logic block diagrams disclosed in the embodiments of the present application.

[0138] In an embodiment of the present application, the processor 1101 in the physiological tissue lesion identification device 1100 will load the instructions corresponding to the processes of one or more applications into the memory 1102 in accordance with the following steps, and the processor 1101 will run the application stored in the memory 1102 to implement various functions. The specific implementation can be found in the previous embodiments and will not be repeated here.

[0139] Optional, such as Figure 8As shown, the physiological tissue lesion identification device 1100 also includes: a touch screen 1103, a radio frequency circuit 1104, an audio circuit 1105, an input unit 1106 and a power supply 1107. Among them, the processor 1101 is electrically connected to the touch screen 1103, the radio frequency circuit 1104, the audio circuit 1105, the input unit 1106 and the power supply 1107 respectively. Those skilled in the art can understand that Figure 8 The structure of the physiological tissue lesion identification device shown in the figure does not constitute a limitation of the physiological tissue lesion identification device, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0140] The touch display screen 1103 can be used to display a graphical user interface and receive operation instructions generated by the user acting on the graphical user interface. The touch display screen 1103 may include a display panel and a touch panel. Among them, the display panel can be used to display information input by the user or information provided to the user and various graphical user interfaces of the physiological tissue lesion identification device, and these graphical user interfaces can be composed of graphics, text, icons, videos and any combination thereof. Optionally, the display panel can be configured in the form of a liquid crystal display (LCD, Liquid Crystal Display), an organic light-emitting diode (OLED, Organic Light-Emitting Diode), etc. The touch panel can be used to collect the user's touch operation on or near it (such as the user uses any suitable object or accessory such as a finger, a stylus, etc. on the touch panel or near the touch panel), and generate corresponding operation instructions, and the operation instructions execute the corresponding program. Optionally, the touch panel may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into the touch point coordinates, and then sends it to the processor 1101, and can receive the command sent by the processor 1101 and execute it. The touch panel can cover the display panel. When the touch panel detects a touch operation on or near it, it is transmitted to the processor 1101 to determine the type of touch event, and then the processor 1101 provides a corresponding visual output on the display panel according to the type of touch event. In an embodiment of the present application, the touch panel and the display panel can be integrated into the touch display screen 1103 to realize the input and output functions. However, in some embodiments, the touch panel and the touch panel can be used as two independent components to realize the input and output functions. That is, the touch display screen 1103 can also be used as a part of the input unit 1106 to realize the input function.

[0141] The radio frequency circuit 1104 can be used to send and receive radio frequency signals to establish wireless communication with network devices or other physiological tissue lesion identification devices through wireless communication, and to send and receive signals between network devices or other physiological tissue lesion identification devices.

[0142] The audio circuit 1105 can be used to provide an audio interface between the user and the physiological tissue lesion identification device through a speaker and a microphone. The audio circuit 1105 can transmit the electrical signal converted from the received audio data to the speaker, which is converted into a sound signal for output; on the other hand, the microphone converts the collected sound signal into an electrical signal, which is received by the audio circuit 1105 and converted into audio data, and then the audio data is output to the processor 1101 for processing, and then sent to another physiological tissue lesion identification device through the radio frequency circuit 1104, or the audio data is output to the memory 1102 for further processing. The audio circuit 1105 may also include an earplug jack to provide communication between an external headset and the physiological tissue lesion identification device.

[0143] The input unit 1106 may be used to receive input numbers, character information or user feature information (such as fingerprint, iris, facial information, etc.), and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0144] The power supply 1107 is used to supply power to various components of the physiological tissue lesion identification device 1100. Optionally, the power supply 1107 can be logically connected to the processor 1101 through a power management device, so as to manage charging, discharging, and power consumption through the power management device. The power supply 1107 can also include one or more DC or AC power supplies, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0145] although Figure 8 Not shown, the physiological tissue lesion identification device 1100 may also include a camera, a sensor, a wireless fidelity module, a Bluetooth module, etc., which will not be repeated here.

[0146] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0147] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0148] To this end, the present application embodiment provides a computer-readable storage medium, in which a plurality of computer programs are stored, and the computer program can be loaded by a processor to execute any of the physiological tissue lesion identification methods provided in the present application embodiment. The computer program can execute the physiological tissue lesion identification method, and the specific implementation can be referred to the previous embodiment, which will not be repeated here.

[0149] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0150] Since the computer program stored in the computer-readable storage medium can execute any of the physiological tissue lesion identification methods provided in the embodiments of the present application, the beneficial effects that can be achieved by any of the physiological tissue lesion identification methods provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0151] According to one aspect of the present application, a computer program product or computer program is also provided, the computer program product or computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. The processor of the physiological tissue lesion identification device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the physiological tissue lesion identification device performs the methods provided in various optional implementations in the above-mentioned embodiments.

[0152] In the above-mentioned physiological tissue lesion identification device, computer-readable storage medium, physiological tissue lesion identification equipment, and computer program product embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process and beneficial effects of the above-described physiological tissue lesion identification device, computer-readable storage medium, computer program product, physiological tissue lesion identification equipment and their corresponding units can refer to the description of the physiological tissue lesion identification method in the above embodiment, and will not be repeated here.

[0153] The above is a detailed introduction to a physiological tissue lesion identification method, device, equipment, readable storage medium and computer program product provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for identifying physiological tissue lesions, characterized in that: The physiological tissue lesion identification method comprises: Acquire surface area images of physiological tissues; Analyzing the surface area image to determine surface features corresponding to the physiological tissue; Based on the surface features and the type of the physiological tissue, a target lesion corresponding to the physiological tissue is determined.

2. The method for identifying physiological tissue lesions according to claim 1, characterized in that: The analyzing the surface area image to determine the surface features corresponding to the physiological tissue includes: Performing smoothing processing on the surface area in the surface area image to obtain a smooth surface area; The surface area is compared with the smooth surface area to determine the corresponding surface features of the physiological tissue; the surface features include the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area.

3. The method for identifying physiological tissue lesions according to claim 1, characterized in that: The analyzing the surface area image to determine the surface features corresponding to the physiological tissue includes: For each pixel point in the surface area image, obtain a first coordinate of the pixel point and a second coordinate of a pixel point adjacent to the pixel point, and determine a direction field of the pixel point according to the first coordinate and the second coordinate; Based on the direction field corresponding to each pixel point in the surface area image, the corresponding surface features of the physiological tissue are determined; the surface features include the roughness of the surface area, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area.

4. The method for identifying physiological tissue lesions according to any one of claims 1 to 3, characterized in that: The determining, based on the surface feature and the type of the physiological tissue, a target lesion corresponding to the physiological tissue comprises: Based on the type of the physiological tissue, query a preset physiological tissue lesion surface feature database to obtain the lesion surface feature corresponding to the type of the physiological tissue; Based on the lesion surface feature and the surface feature, a target lesion corresponding to the physiological tissue is determined.

5. The method for identifying physiological tissue lesions according to claim 4, characterized in that: The determining, based on the lesion surface feature and the surface feature, a target lesion corresponding to the physiological tissue comprises: The roughness of the surface area in the surface feature, the distribution density of the non-smooth area of ​​the surface area, and the area of ​​the non-smooth area are counted and quantified to determine a first evaluation index corresponding to the surface feature; The unevenness of the lesion surface area, the distribution density of the non-smooth area of ​​the lesion surface area, and the area of ​​the non-smooth area in each of the lesion surface features are counted and quantified to determine a second evaluation index corresponding to the lesion surface feature; The similarity between the first evaluation indicator and each of the second evaluation indicators is calculated respectively, a target evaluation indicator is determined based on the similarity, and the lesion corresponding to the target evaluation indicator is determined as the target lesion corresponding to the physiological tissue, and the target evaluation indicator is the second evaluation indicator having the maximum similarity with the first evaluation indicator and the similarity is greater than a preset similarity threshold.

6. The method for identifying physiological tissue lesions according to claim 1, characterized in that: The step of acquiring a surface area image of a physiological tissue comprises: Acquire the type and image of physiological tissue; Selecting a target segmentation model from a preset segmentation model library based on the type of the physiological tissue, and segmenting the physiological tissue image based on the target segmentation model to obtain a physiological tissue segmentation image; A surface area image of the physiological tissue is extracted from the physiological tissue segmentation image.

7. The method for identifying physiological tissue lesions according to claim 1, characterized in that: Before analyzing the surface area image to determine the surface features corresponding to the physiological tissue, the method includes: Constructing a buffer zone in the surface area image, and calculating edge features of each pixel point in the buffer zone; The clarity of the surface area boundary of the surface area image is optimized based on the edge feature to obtain an optimized surface area image.

8. A physiological tissue lesion identification device, characterized in that: The physiological tissue lesion identification device comprises: An acquisition module, used for acquiring a surface area image of a physiological tissue; An analysis module, used for analyzing the surface area image to determine the surface features corresponding to the physiological tissue; A determination module is used to determine a target lesion corresponding to the physiological tissue based on the surface feature and the type of the physiological tissue.

9. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores a plurality of computer programs, and the processor loads the computer programs from the memory to execute the steps of the physiological tissue lesion identification method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of computer programs, and the computer programs are suitable for being loaded by a processor to execute the steps of the physiological tissue lesion identification method according to any one of claims 1-7.

Images