Peripheral blood vessel image auxiliary segmentation method and system

Abstract

The invention relates to the technical field of peripheral blood vessel image segmentation, in particular to a peripheral blood vessel image auxiliary segmentation method and system. The method comprises the following steps: obtaining a lesion intensity index according to gray distribution and edge distribution in a preset neighborhood of lesion pixels; obtaining a shape feature factor according to the contour feature of the lesion area; obtaining a structural characteristic factor according to the gray distribution of the lesion area; obtaining a range characteristic factor according to the number of pixel points in the lesion area; obtaining an expansion weight coefficient according to the characteristic factor difference and the lesion intensity index; obtaining an expansion sampling priority degree according to the expansion weight coefficient distribution of the lesion area in each lesion blood vessel image; further obtaining an expanded training model; and segmenting the peripheral blood vessel image according to the expanded training model. According to the method, the expanded peripheral blood vessel sample is balanced and abundant, the generalization ability of a deep learning segmentation network is improved, and the accuracy of auxiliary segmentation is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of peripheral blood vessel image segmentation, and in particular to a peripheral blood vessel image auxiliary segmentation method and system. Background Art

[0002] Peripheral blood vessels refer to the blood vessels in the human body's limbs, head, neck, trunk, etc., except for the cardiovascular and cerebrovascular vessels. Auxiliary segmentation of images of human peripheral blood vessels has important clinical and research significance. It can help doctors and researchers accurately locate the position and morphology of vascular structures, thereby providing certain support for diagnosing diseases, formulating treatment plans, and studying changes in vascular anatomical structures, saving a lot of time and labor costs.

[0003] In the prior art, deep learning segmentation networks are often used to assist in the segmentation of peripheral vascular images. Since various types of lesions may occur in the peripheral blood vessels of the human body, a large number of lesion samples are required to train the deep learning segmentation network when the deep learning segmentation network is used for auxiliary segmentation, so the peripheral vascular images need to be expanded. However, the prior art usually expands peripheral vascular images by random sampling, without considering the types, structures, and distributions of peripheral vascular lesions. As a result, the expanded peripheral blood vessels will have unbalanced samples and insufficient richness, which reduces the generalization ability of the deep learning segmentation network and makes it difficult to better segment various types of peripheral vascular images. Summary of the invention

[0004] In order to solve the technical problem that the prior art usually expands peripheral vascular images by random sampling without considering the different types, structures, and distributions of peripheral vascular lesions, resulting in sample imbalance and insufficient richness of the expanded peripheral blood vessels, reducing the generalization ability of the deep learning segmentation network, and making it difficult to segment various types of peripheral vascular images well, the purpose of the present invention is to provide a peripheral vascular image-assisted segmentation method and system, and the technical scheme adopted is as follows: a peripheral vascular image-assisted segmentation method, the method comprising: collecting all peripheral vascular images; taking the peripheral vascular image with lesions as the lesion vascular image; selecting any lesion pixel point in any lesion vascular image as a reference lesion pixel point; obtaining the lesion intensity index of the reference lesion pixel point according to the grayscale distribution in a preset neighborhood of the reference lesion pixel point and the gradient distribution of the peripheral vascular edge pixel point; and obtaining the lesion intensity index of the reference lesion pixel point according to each lesion vessel. The lesion pixel distribution of the image is used to obtain the lesion area in each lesion blood vessel image; one lesion area is selected as the reference lesion area; according to the contour characteristics of the reference lesion area, the shape characteristic factor of the reference lesion area is obtained; according to the grayscale distribution in the reference lesion area, the structural characteristic factor of the reference lesion area is obtained; according to the number characteristics of lesion pixels in the reference lesion area, the range characteristic factor of the reference lesion area is obtained; according to the shape characteristic factor difference, structural characteristic factor difference, range characteristic factor difference of the reference lesion area compared with other lesion areas and the lesion intensity index of each lesion pixel, the expansion weight coefficient of the reference lesion area is obtained; according to the expansion weight coefficient distribution characteristics of all lesion areas in each lesion blood vessel image, the expansion sampling priority of each lesion blood vessel image is obtained; based on the expansion sampling priority, an expanded training model is obtained; and the peripheral blood vessel image is segmented according to the expanded training model.

[0005] Furthermore, the method for obtaining the lesion strength index includes: obtaining the lesion strength index according to a lesion strength index calculation formula, and the lesion strength index calculation formula is as follows: In the formula, represents the lesion intensity index of the reference lesion pixel; Represents the number of pixels within a preset neighborhood of a reference lesion pixel; Indicates the first pixel in the preset neighborhood of the reference lesion pixel. The gray value of each pixel; Represents the grayscale mean of the pixels within the preset neighborhood of the reference lesion pixel; Represents the grayscale mean of the pixel points in the lesion vessel image where the reference lesion pixel point is located; Represents the number of peripheral blood vessel edge pixels within a preset neighborhood of a reference lesion pixel; Represents the first The gradient direction angle of the peripheral blood vessel edge pixel point; Represents the mean value of the gradient direction angle of the peripheral blood vessel edge pixel points within the preset neighborhood of the reference lesion pixel point.

[0006] Furthermore, the method for obtaining the lesion area includes: taking the pixel points corresponding to the peripheral blood vessel area where the lesion occurs in each lesion blood vessel image as the lesion pixel points; using a connected domain labeling algorithm to obtain all connected domains formed by all lesion pixel points in each lesion blood vessel image, and taking all the connected domains as all lesion areas in each lesion blood vessel image.

[0007] Furthermore, the method for obtaining the shape characteristic factor includes: obtaining all edge pixel points and centroid pixel points of the reference lesion area; obtaining the line between the centroid pixel point and all edge pixel points as a shape characteristic line segment; taking the centroid pixel point as the center, equally dividing the reference lesion area along a preset angle to obtain all lesion modules; selecting the longest shape characteristic line segment and the shortest shape characteristic line segment in each lesion module, and taking the angle formed by each adjacent two longest shape characteristic line segments of all lesion modules as the first shape characteristic angle, and taking the angle formed by each adjacent two shortest shape characteristic line segments of all lesion modules as the second shape characteristic angle; according to the first shape characteristic angle and the second shape characteristic angle, obtaining the shape characteristic factor of the reference lesion area, the calculation formula is as follows: In the formula, represents the shape characteristic factor of the reference lesion area; represents the number of first shape feature angles and the number of second shape feature angles; represents the first-order difference of all first shape feature angles; represents the first-order difference of all the second shape feature angles; represents the absolute value function.

[0008] Furthermore, the method for obtaining the structural characteristic factor includes: obtaining the structural characteristic factor according to a structural characteristic factor calculation formula, and the structural characteristic factor calculation formula is as follows: In the formula, represents the structural characteristic factor of the reference lesion area; Indicates the number of shape feature line segments in the reference lesion area; Indicates the first The grayscale mean of the pixels in the bar-shaped feature segment; Indicates the first The grayscale mean of the pixels in the bar-shaped feature segment; Indicates the first The number of pixels in the bar shape feature segment; Indicates the first The first of the shape feature segments The gray value of each pixel; Indicates the first The first of the shape feature segments The gray value of each pixel; represents the absolute value function.

[0009] Furthermore, the method for obtaining the range characteristic factor includes: obtaining the range characteristic factor according to a range characteristic factor calculation formula, and the range characteristic factor calculation formula is as follows: In the formula, represents the range characteristic factor of the reference lesion area; represents the number of pixels in the reference lesion area; Indicates the number of pixels in all lesion areas in the lesion vessel image where the reference lesion area is located; Represents the number of pixels in the lesion vessel image where the reference lesion area is located.

[0010] Furthermore, the method for obtaining the expanded weight coefficient includes: obtaining the expanded weight coefficient according to an expanded weight coefficient calculation formula, and the expanded weight coefficient calculation formula is as follows: In the formula, represents the expansion weight coefficient of the reference lesion area; represents the number of pixels in the reference lesion area; Indicates the first The lesion intensity index of each pixel; Indicates the number of other lesion areas in the lesion vessel image where the reference lesion area is located; represents the shape characteristic factor of the reference lesion area; Indicates the first Shape characteristic factors of other lesion areas; represents the structural characteristic factor of the reference lesion area; Indicates the first structural characteristic factors of other lesion areas; represents the range characteristic factor of the reference lesion area; Indicates the first Other range characteristics of the lesion area; represents the absolute value function.

[0011] Furthermore, the method for obtaining the expanded sampling priority includes: taking any diseased blood vessel image as the target image; accumulating and summing the lesion intensity indices of all lesion areas in the target image to obtain a target weight coefficient of the target image; accumulating and summing the lesion intensity indices of all lesion areas in all lesion blood vessel images to obtain a total image weight coefficient; taking the ratio of the target weight coefficient to the total image weight coefficient as the expanded sampling priority of the target image; and changing the target image to obtain the expanded sampling priority of each lesion blood vessel image.

[0012] Furthermore, based on the expansion sampling priority, an expanded training model is obtained, including: taking a peripheral vascular image that does not include a lesion area as a normal vascular image; enhancing and expanding all acquired lesion vascular images according to the expansion sampling priority of all lesion vascular images, until the expanded lesion vascular images reach a preset first number, and taking the expanded lesion vascular images and normal vascular images as the expanded training model.

[0013] A peripheral vascular image-assisted segmentation system comprises a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned peripheral vascular image-assisted segmentation method are implemented.

[0014] The present invention has the following beneficial effects: the present invention collects all peripheral vascular images, in order to facilitate the subsequent segmentation of peripheral vascular images using a deep learning segmentation network; since there are differences in grayscale, structure, and peripheral vascular edges between the lesion area and other normal areas, the grayscale distribution in the preset neighborhood of the lesion pixel point and the gradient distribution of the peripheral vascular edge pixel point are used to obtain the lesion intensity index of the lesion pixel point; since different types of peripheral vascular lesions will cause differences in shape, structure, and range of peripheral blood vessels, the shape characteristic factors, structural characteristic factors, and range characteristic factors of each lesion area are analyzed; in order to improve the deep learning To improve the segmentation ability of the deep learning segmentation network, it is necessary to focus on the lesion areas with higher overall lesion intensity index and lower occurrence frequency for expansion. Therefore, the expansion weight coefficient of the reference lesion area is obtained according to the difference in shape feature factors, structural feature factors, range feature factors and lesion intensity index of each lesion pixel point between the reference lesion area and other lesion areas; since the lesion areas in each lesion vessel image have different importance when expanding the training samples, the expansion sampling priority of each lesion vessel image is analyzed; the expanded training model is obtained based on the expansion sampling priority; the peripheral vascular image is segmented according to the expanded training model. The present invention takes into account the characteristics of different peripheral vascular lesions, resulting in balanced and rich peripheral vascular samples after expansion, increasing the generalization ability of the deep learning segmentation network and improving the accuracy of auxiliary segmentation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A flow chart of a peripheral vascular image-assisted segmentation method provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the peripheral vascular image-assisted segmentation method and system proposed by the present invention, its specific implementation, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The specific scheme of a peripheral blood vessel image-assisted segmentation method and system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a peripheral vascular image-assisted segmentation method provided by an embodiment of the present invention, the method comprising: step S1: acquiring all peripheral vascular images; and taking the peripheral vascular images with lesions as lesion vascular images.

[0021] The embodiment of the present invention is mainly used in the scenario of sample expansion of peripheral vascular images. In order to expand the peripheral vascular images, the peripheral vascular images are first obtained, and the samples of the peripheral vascular images with lesions are expanded. Since the deep learning segmentation network is used to segment the peripheral vascular images in the embodiment of the present invention, and the deep learning segmentation network requires a large number of training samples, that is, peripheral vascular images containing lesion areas, the peripheral vascular images with lesions are used as lesion vascular images, and the lesion vascular images are subsequently enhanced and expanded to ensure that the deep learning segmentation network has enough training samples.

[0022] In one embodiment of the present invention, a U-Net deep learning model is used to assist in segmenting peripheral vascular images. It should be noted that in other embodiments of the present invention, models such as convolutional neural networks (CNNs) may also be used to perform the same operation, and both the U-Net deep learning model and the convolutional neural network are well known to those skilled in the art, and are not limited or elaborated herein.

[0023] In one embodiment of the present invention, since there is a large difference between the grayscale values ​​of pixels in the diseased area of ​​the peripheral blood vessels and the grayscale values ​​of other normal areas, and the area of ​​the peripheral blood vessel image occupied by the diseased area is small, an abnormal label is manually added to the pixels in the diseased area, and the peripheral blood vessel image with the abnormal label is used as the diseased blood vessel image.

[0024] Step S2: Select any lesion pixel point in any lesion blood vessel image as a reference lesion pixel point; obtain the lesion intensity index of the reference lesion pixel point according to the grayscale distribution in a preset neighborhood of the reference lesion pixel point and the gradient distribution of the peripheral blood vessel edge pixel points; obtain the lesion area in each lesion blood vessel image according to the lesion pixel point distribution of each lesion blood vessel image; select any lesion area as a reference lesion area; obtain the shape characteristic factor of the reference lesion area according to the contour characteristics of the reference lesion area; obtain the structural characteristic factor of the reference lesion area according to the grayscale distribution in the reference lesion area; obtain the range characteristic factor of the reference lesion area according to the number characteristics of the lesion pixels in the reference lesion area.

[0025] The peripheral blood vessels of the human body may undergo pathological changes due to various reasons, such as deformation, distortion, and blockage of the peripheral blood vessels. The segmentation effects of diseased blood vessel images containing different degrees of lesion areas are different. In order to perform better segmentation of all types of peripheral vascular lesions, it is necessary to provide enough training samples for each type of peripheral vascular lesions in the U-Net deep learning model. In order to describe different types of peripheral vascular lesions, it is necessary to calculate the intensity of the pixels in the lesion area in each lesion blood vessel image. Since there is a large difference between the grayscale value of the lesion area and the grayscale value of other normal areas, and different types of peripheral vascular lesions will lead to large differences in the corresponding peripheral vascular structures, that is, different types of peripheral vascular lesions will lead to differences in the edges of the peripheral blood vessels. Therefore, in an embodiment of the present invention, the lesion intensity index of the reference lesion pixel is obtained according to the grayscale distribution in a preset neighborhood of the reference lesion pixel and the gradient distribution of the peripheral blood vessel edge pixel.

[0026] Preferably, in one embodiment of the present invention, a method for acquiring a lesion region comprises: marking the peripheral vascular region where lesions occur in each lesion vessel image according to the prior art, and taking the pixel points corresponding to the peripheral vascular region where lesions occur in each lesion vessel image as lesion pixel points; and obtaining all connected domains formed by all lesion pixel points in each lesion vessel image using a connected domain marking algorithm, and taking all connected domains as all lesion regions in each lesion vessel image.

[0027] Preferably, in one embodiment of the present invention, the method for obtaining the lesion intensity index includes: obtaining the lesion intensity index according to a lesion intensity index calculation formula, and the lesion intensity index calculation formula is as follows: In the formula, represents the lesion intensity index of the reference lesion pixel; Represents the number of pixels within a preset neighborhood of a reference lesion pixel; Indicates the first pixel in the preset neighborhood of the reference lesion pixel. The gray value of each pixel; Represents the grayscale mean of the pixels within the preset neighborhood of the reference lesion pixel; Represents the grayscale mean of the pixel points in the lesion vessel image where the reference lesion pixel point is located; Represents the number of peripheral blood vessel edge pixels within a preset neighborhood of a reference lesion pixel; Represents the first The gradient direction angle of the peripheral blood vessel edge pixel point, wherein the gradient direction angle of the peripheral blood vessel edge pixel point can be obtained by the existing technology and will not be described in detail here; Represents the mean value of the gradient direction angle of the peripheral blood vessel edge pixel points within the preset neighborhood of the reference lesion pixel point.

[0028] In the lesion intensity index calculation formula, the difference between the gray value of each pixel in the preset neighborhood of the reference pixel and the overall gray value of the preset neighborhood The larger the value, the more uneven the blood flow around the reference pixel. At this time, the area where the reference pixel is located is more likely to have a lesion, that is, the greater the lesion intensity index of the reference pixel. The difference between the grayscale value of each pixel in the preset neighborhood of the reference pixel and the overall grayscale value of the lesion vessel image is. The larger the value is, the more significant the grayscale value of the pixel in the preset neighborhood is in the lesion blood vessel image, that is, the greater the difference between the blood flow in the preset neighborhood where the reference pixel is located and the blood flow in other areas, that is, the more likely the area where the reference pixel is located is to have a lesion, that is, the greater the lesion intensity index of the reference pixel; The larger it is, the greater the change in the gradient direction of the edge pixel points of the peripheral blood vessels, that is, the peripheral blood vessels where the reference pixel points are located are more likely to undergo structural changes. At this time, the lesion intensity index of the peripheral blood vessel area corresponding to the reference pixel points is greater.

[0029] In one embodiment of the present invention, the reference pixel point is taken as the center. The rectangular area is set as the preset neighborhood of the reference pixel. It should be noted that in other embodiments of the present invention, the preset neighborhood can be set voluntarily, which is not limited here.

[0030] In reality, different types of peripheral vascular diseases will cause differences in shape, structure, and range of peripheral blood vessels. In order to ensure that the number of samples of various types of peripheral vascular diseases remains stable when expanding the training samples of the deep learning segmentation network, in an embodiment of the present invention, the shape characteristics, structural characteristics, and range characteristics of each lesion area are analyzed.

[0031] Since the lesion areas of peripheral blood vessels usually have large differences in shape and may appear elliptical, straight, branched, etc., in one embodiment of the present invention, any lesion area is selected as a reference lesion area, and the shape characteristic factor of the reference lesion area is obtained based on the contour characteristics of the reference lesion area.

[0032] Preferably, in one embodiment of the present invention, the method for acquiring the shape characteristic factor includes: acquiring all edge pixel points and centroid pixel points of the reference lesion area.

[0033] The connecting line between the centroid pixel and all edge pixels is obtained as the shape feature line segment.

[0034] Taking the centroid pixel point as the center, the reference lesion area is equally divided along the preset angle to obtain all lesion modules. In one embodiment of the present invention, the preset angle is set to 30°, that is, taking the centroid pixel point as the center, the longest shape feature line segment in the reference lesion area is rotated, and the reference lesion area through which the longest shape feature line segment passes when rotating 30° is used as each lesion module, so that 12 lesion modules can be obtained from the reference lesion area. It should be noted that in other embodiments of the present invention, the preset angle can be set voluntarily and is not limited here.

[0035] In each lesion module, the longest shape characteristic line segment and the shortest shape characteristic line segment are selected, and the angle formed by every two adjacent longest shape characteristic line segments of all lesion modules is taken as the first shape characteristic angle, and the angle formed by every two adjacent shortest shape characteristic line segments of all lesion modules is taken as the second shape characteristic angle; according to the above process, it can be known that in one embodiment of the present invention, the number of first shape characteristic angles and second shape characteristic angles is the same, and there are 11 of each.

[0036] According to the first shape characteristic angle and the second shape characteristic angle, the shape characteristic factor of the reference lesion area is obtained, and the calculation formula is as follows: In the formula, represents the shape characteristic factor of the reference lesion area; represents the number of first shape feature angles and the number of second shape feature angles; a first-order difference of angles representing all first shape feature angles; a first-order difference of angles representing all included angles of the second shape features; represents the absolute value function.

[0037] In the shape feature factor calculation formula, the angle formed by the adjacent longest shape feature line segments of different lesion modules can reflect the contour feature distribution of the edge of the reference lesion area far away from the centroid pixel point, and the angle formed by the adjacent shortest shape feature line segments can reflect the contour feature distribution of the reference lesion area close to the centroid pixel point; the angle formed by all adjacent longest shape feature line segments, that is, the first-order difference of the angle of the first shape feature angle, and the angle formed by the adjacent shortest shape feature line segments, that is, the first-order difference of the angle of the second shape feature angle, are averaged to obtain the shape feature factor of the reference lesion area.

[0038] In reality, pathological changes in peripheral blood vessels may cause cysts, calcifications, inflammation, etc. At this time, the tissue composition density of the peripheral blood vessels is different, which is reflected in the peripheral blood vessel image as different grayscale distributions of the lesion area. Therefore, in an embodiment of the present invention, the structural characteristic factor of the reference lesion area is obtained based on the grayscale distribution in the reference lesion area.

[0039] Preferably, in one embodiment of the present invention, the method for obtaining the structural characteristic factor includes: obtaining the structural characteristic factor according to a structural characteristic factor calculation formula, and the structural characteristic factor calculation formula is as follows: In the formula, represents the structural characteristic factor of the reference lesion area; Indicates the number of shape feature line segments in the reference lesion area; Indicates the first The grayscale mean of the pixels in the bar-shaped feature segment; Indicates the first The grayscale mean of the pixels in the bar-shaped feature segment; Indicates the first The number of pixels in the bar shape feature segment; Indicates the first The first of the shape feature segments The gray value of each pixel; Indicates the first The first of the shape feature segments The gray value of each pixel; represents the absolute value function.

[0040] In the calculation formula of the structural characteristic factor, the average grayscale difference between two adjacent shape characteristic line segments in the reference lesion area is averaged. The larger the value, the greater the overall grayscale change of the reference lesion area; the grayscale difference between each two adjacent pixels in each shape feature line segment of the reference lesion area is averaged. , where the larger the mean, the greater the grayscale change from the centroid pixel to the edge of the reference lesion area; so As the structural characteristic factor of the reference lesion area.

[0041] Since the areas of different types of peripheral vascular lesions may vary greatly, in an embodiment of the present invention, a range characteristic factor of the reference lesion area is obtained based on the number characteristics of lesion pixels in the reference lesion area.

[0042] Preferably, in one embodiment of the present invention, the method for obtaining the range characteristic factor includes: obtaining the range characteristic factor according to a range characteristic factor calculation formula, and the range characteristic factor calculation formula is as follows: In the formula, represents the range characteristic factor of the reference lesion area; represents the number of pixels in the reference lesion area; Indicates the number of pixels in all lesion areas in the lesion vessel image where the reference lesion area is located; Represents the number of pixels in the lesion vessel image where the reference lesion area is located.

[0043] In the range characteristic factor calculation formula, the ratio of the number of pixels in the reference lesion area to all lesion areas in the peripheral vascular image is calculated. , and calculate the ratio between the number of pixels in the reference lesion area and the number of all pixels in the peripheral vascular image The larger the ratio, the more prominent the reference lesion area is in the range of all lesion areas. As the range characteristic factor of the reference lesion area.

[0044] Step S3: Obtain the expansion weight coefficient of the reference lesion area according to the difference in shape characteristic factors, structural characteristic factors, range characteristic factors and the lesion intensity index of each lesion pixel between the reference lesion area and other lesion areas; obtain the expansion sampling priority of each lesion vessel image according to the distribution characteristics of the expansion weight coefficients of all lesion areas in each lesion vessel image.

[0045] According to the above process, the shape characteristic factor, structure characteristic factor, range characteristic factor and lesion intensity index of each pixel in the lesion area are obtained for each lesion area; in order to improve the segmentation ability of the deep learning segmentation network, it is necessary to focus on the lesion areas with higher overall lesion intensity index and lower occurrence frequency when expanding the samples, so as to enhance the balance and richness of the final samples in the deep learning segmentation network. Therefore, the expansion weight coefficient of the reference lesion area is obtained according to the difference in shape characteristic factor, structure characteristic factor and range characteristic factor between the reference lesion area and other lesion areas, as well as the lesion intensity index of each lesion pixel.

[0046] Preferably, in one embodiment of the present invention, the method for obtaining the expanded weight coefficient includes: obtaining the expanded weight coefficient according to an expanded weight coefficient calculation formula, and the expanded weight coefficient calculation formula is as follows: In the formula, represents the expansion weight coefficient of the reference lesion area; represents the number of pixels in the reference lesion area; Indicates the first The lesion intensity index of each pixel; Indicates the number of other lesion areas in the lesion vessel image where the reference lesion area is located; represents the shape characteristic factor of the reference lesion area; Indicates the first Shape characteristic factors of other lesion areas; represents the structural characteristic factor of the reference lesion area; Indicates the first structural characteristic factors of other lesion areas; represents the range characteristic factor of the reference lesion area; Indicates the first Other range characteristics of the lesion area; represents the absolute value function.

[0047] In the expanded weight coefficient calculation formula, the total lesion intensity index of the reference lesion area is calculated , where the greater the total lesion intensity index, the more serious the lesion in the reference area. At this time, the reference lesion area should be paid more attention when the training sample is expanded, that is, the higher the weight coefficient of the reference lesion area when expanding; calculate the difference in shape feature factor, structure feature factor difference, and range feature factor difference between the reference lesion area and all other lesion areas and calculate the average , where the larger the difference mean, the more unique the lesion in the reference lesion area, and the lower the frequency of peripheral vascular lesions corresponding to the reference lesion area. At this time, it is more necessary to consider the uniqueness of the reference lesion area when expanding the training samples, that is, the higher the weight coefficient of the reference lesion area when expanding.

[0048] Since the importance of the lesion area in each lesion vessel image when expanding the training samples is different, it is necessary to analyze the expansion weight coefficients of all lesion areas in each lesion vessel image when expanding the samples. That is, in the embodiment of the present invention, the expansion sampling priority of each lesion vessel image is obtained according to the distribution characteristics of the expansion weight coefficients of all lesion areas in each lesion vessel image, so as to pay attention to more important lesion vessel images in the future.

[0049] Preferably, in one embodiment of the present invention, a method for obtaining an expanded sampling priority includes: taking any diseased blood vessel image as a target image; accumulating and summing the lesion intensity indices of all lesion areas in the target image to obtain a target weight coefficient of the target image; accumulating and summing the lesion intensity indices of all lesion areas in all lesion blood vessel images to obtain a total image weight coefficient; taking a ratio between the target weight coefficient and the total image weight coefficient as an expanded sampling priority of the target image; and changing the target image to obtain an expanded sampling priority of each lesion blood vessel image.

[0050] In one embodiment of the present invention, the calculation formula for the expanded sampling priority includes: In the formula, Indicates the priority of the target image's expansion sampling; Indicates the number of lesion areas in the target image; Indicates the target image The expansion weight coefficient of each lesion area; represents the number of lesion areas in all lesion vessel images; Indicates the first The expansion weight coefficient of the lesion area.

[0051] In the expansion sampling priority calculation formula, the greater the ratio of the sum of the expansion weight coefficients of all lesion areas of the target image to the sum of the expansion weight coefficients of the lesion areas in all lesion vessel images, the more important the target image is, and the greater the priority of being extracted when the training samples are expanded. At this time, more samples should be taken for the target image when the training samples are enhanced and expanded, that is, the greater the expansion sampling priority.

[0052] Step S4: obtaining an expanded training model based on the expanded sampling priority; and segmenting the peripheral vascular image according to the expanded training model.

[0053] Preferably, in one embodiment of the present invention, obtaining an expanded training model based on the expanded sampling priority includes: taking a peripheral vascular image that does not contain a lesion area as a normal vascular image; calculating the ratio between the expanded sampling priority of each lesion vascular image and the expanded sampling priority of all lesion vascular images as a first ratio, sampling and expanding all lesion vascular images so that the ratio between each type of newly added training samples meets the first ratio, until the expanded lesion vascular images reach a preset first number, and taking the expanded lesion vascular images and normal vascular images as all expanded peripheral vascular images. In one embodiment of the present invention, the preset first number is set to 25% of all expanded peripheral vascular images. It should be noted that in other embodiments of the present invention, the preset first number can be set voluntarily and is not limited here.

[0054] The U-Net model is trained using all the expanded peripheral vascular images to obtain a model with good segmentation effect on various types of diseased peripheral vascular images as the expanded training model. The expanded training model is used to segment the peripheral vascular images to avoid the problem of incomplete segmentation results caused by vascular lesions to the greatest extent, improve the accuracy of auxiliary segmentation, and complete the auxiliary segmentation of peripheral vascular images.

[0055] At this point, the segmentation of the peripheral blood vessel image is completed.

[0056] In summary, all peripheral vascular images are collected; peripheral vascular images with lesions are taken as lesion vascular images; any lesion pixel point in any lesion vascular image is selected as a reference lesion pixel point; the lesion intensity index of the reference lesion pixel point is obtained according to the grayscale distribution in the preset neighborhood of the reference lesion pixel point and the gradient distribution of the peripheral vascular edge pixel points; the lesion area in each lesion vascular image is obtained according to the lesion pixel distribution of each lesion vascular image; any lesion area is selected as the reference lesion area; the shape characteristic factor of the reference lesion area is obtained according to the contour characteristics of the reference lesion area; the grayscale distribution in the reference lesion area is obtained. The structural characteristic factor of the reference lesion area is obtained; according to the number characteristics of the lesion pixels in the reference lesion area, the range characteristic factor of the reference lesion area is obtained; according to the difference in shape characteristic factors, structural characteristic factors, range characteristic factors and the lesion intensity index of each lesion pixel between the reference lesion area and other lesion areas, the expansion weight coefficient of the reference lesion area is obtained; according to the distribution characteristics of the expansion weight coefficients of all lesion areas in each lesion vessel image, the expansion sampling priority of each lesion vessel image is obtained; based on the expansion sampling priority, an expanded training model is obtained; and the peripheral vessel image is segmented according to the expanded training model.

[0057] An embodiment of the present invention also provides a peripheral vascular image-assisted segmentation system, which includes a memory, a processor and a computer program, wherein the memory is used to store the corresponding computer program, and the processor is used to run the corresponding computer program. When the computer program runs in the processor, it can implement the method described in steps S1-S4.

[0058] It should be noted that the sequence of the above embodiments of the present invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0059] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A peripheral vascular image-assisted segmentation method, characterized in that: The method comprises: collecting all peripheral vascular images; taking the peripheral vascular images with lesions as lesion vascular images; selecting a lesion pixel point in any lesion vascular image as a reference lesion pixel point; obtaining a lesion intensity index of the reference lesion pixel point according to the grayscale distribution in a preset neighborhood of the reference lesion pixel point and the gradient distribution of the peripheral vascular edge pixel points; obtaining the lesion area in each lesion vascular image according to the lesion pixel point distribution of each lesion vascular image; selecting a lesion area as a reference lesion area; obtaining a shape feature factor of the reference lesion area according to the contour features of the reference lesion area; obtaining a shape feature factor of the reference lesion area according to the grayscale distribution in the reference lesion area. Obtain a structural characteristic factor of a reference lesion area; obtain a range characteristic factor of the reference lesion area according to the number characteristics of lesion pixels in the reference lesion area; obtain an expansion weight coefficient of the reference lesion area according to the difference in shape characteristic factors, structural characteristic factors, range characteristic factors and the lesion intensity index of each lesion pixel between the reference lesion area and other lesion areas; obtain an expansion sampling priority of each lesion vessel image according to the distribution characteristics of the expansion weight coefficients of all lesion areas in each lesion vessel image; obtain an expanded training model based on the expanded sampling priority; and segment the peripheral vessel image according to the expanded training model.

2. The peripheral vascular image-assisted segmentation method according to claim 1, characterized in that: The method for obtaining the lesion intensity index includes: obtaining the lesion intensity index according to a lesion intensity index calculation formula, and the lesion intensity index calculation formula is as follows: In the formula, represents the lesion intensity index of the reference lesion pixel; Represents the number of pixels within a preset neighborhood of a reference lesion pixel; Indicates the first pixel in the preset neighborhood of the reference lesion pixel. The gray value of each pixel; Represents the grayscale mean of the pixels within the preset neighborhood of the reference lesion pixel; Represents the grayscale mean of the pixel points in the lesion vessel image where the reference lesion pixel point is located; Represents the number of peripheral blood vessel edge pixels within a preset neighborhood of a reference lesion pixel; Represents the first The gradient direction angle of the peripheral blood vessel edge pixel point; Represents the mean value of the gradient direction angle of the peripheral blood vessel edge pixel points within the preset neighborhood of the reference lesion pixel point.

3. The peripheral vascular image-assisted segmentation method according to claim 1, characterized in that: The method for acquiring the lesion area includes: taking the pixel points corresponding to the peripheral blood vessel area where the lesion occurs in each lesion blood vessel image as the lesion pixel points; using a connected domain labeling algorithm to obtain all connected domains formed by all lesion pixel points in each lesion blood vessel image, and taking all the connected domains as all lesion areas in each lesion blood vessel image.

4. The peripheral vascular image-assisted segmentation method according to claim 1, characterized in that: The method for obtaining the shape characteristic factor includes: obtaining all edge pixel points and centroid pixel points of the reference lesion area; obtaining the connection line between the centroid pixel point and all edge pixel points as a shape characteristic line segment; taking the centroid pixel point as the center, equally dividing the reference lesion area along a preset angle to obtain all lesion modules; selecting the longest shape characteristic line segment and the shortest shape characteristic line segment in each lesion module, and taking the angle formed by each adjacent two longest shape characteristic line segments of all lesion modules as the first shape characteristic angle, and taking the angle formed by each adjacent two shortest shape characteristic line segments of all lesion modules as the second shape characteristic angle; according to the first shape characteristic angle and the second shape characteristic angle, the shape characteristic factor of the reference lesion area is obtained, and the calculation formula is as follows: In the formula, represents the shape characteristic factor of the reference lesion area; represents the number of first shape feature angles and the number of second shape feature angles; represents the first-order difference of all first shape feature angles; represents the first-order difference of all the second shape feature angles; represents the absolute value function.

5. The peripheral vascular image-assisted segmentation method according to claim 4, characterized in that: The method for obtaining the structural characteristic factor includes: obtaining the structural characteristic factor according to a structural characteristic factor calculation formula, and the structural characteristic factor calculation formula is as follows: In the formula, represents the structural characteristic factor of the reference lesion area; Indicates the number of shape feature line segments in the reference lesion area; Indicates the first The grayscale mean of the pixels in the bar-shaped feature segment; Indicates the first The grayscale mean of the pixels in the bar-shaped feature segment; Indicates the first The number of pixels in the bar shape feature segment; Indicates the first The first of the shape feature segments The gray value of each pixel; Indicates the first The first of the shape feature segments The gray value of each pixel; represents the absolute value function.

6. The peripheral vascular image-assisted segmentation method according to claim 1, characterized in that: The method for obtaining the range characteristic factor includes: obtaining the range characteristic factor according to a range characteristic factor calculation formula, and the range characteristic factor calculation formula is as follows: In the formula, represents the range characteristic factor of the reference lesion area; represents the number of pixels in the reference lesion area; Indicates the number of pixels in all lesion areas in the lesion vessel image where the reference lesion area is located; Represents the number of pixels in the lesion vessel image where the reference lesion area is located.

7. The peripheral vascular image-assisted segmentation method according to claim 1, characterized in that: The method for obtaining the expanded weight coefficient includes: obtaining the expanded weight coefficient according to an expanded weight coefficient calculation formula, and the expanded weight coefficient calculation formula is as follows: In the formula, represents the expansion weight coefficient of the reference lesion area; represents the number of pixels in the reference lesion area; Indicates the first The lesion intensity index of each pixel; Indicates the number of other lesion areas in the lesion vessel image where the reference lesion area is located; represents the shape characteristic factor of the reference lesion area; Indicates the first Shape characteristic factors of other lesion areas; represents the structural characteristic factor of the reference lesion area; Indicates the first structural characteristic factors of other lesion areas; represents the range characteristic factor of the reference lesion area; Indicates the first Other range characteristics of the lesion area; represents the absolute value function.

8. The peripheral vascular image-assisted segmentation method according to claim 1, characterized in that: The method for obtaining the extended sampling priority includes: taking any diseased blood vessel image as a target image; accumulating and summing the lesion intensity indices of all lesion areas in the target image to obtain a target weight coefficient of the target image; accumulating and summing the lesion intensity indices of all lesion areas in all lesion blood vessel images to obtain a total image weight coefficient; taking the ratio between the target weight coefficient and the total image weight coefficient as the extended sampling priority of the target image; and changing the target image to obtain the extended sampling priority of each lesion blood vessel image.

9. The peripheral vascular image-assisted segmentation method according to claim 1, characterized in that: The expanded training model is obtained based on the expansion sampling priority, including: taking the peripheral vascular image that does not include the lesion area as the normal vascular image; enhancing and expanding all the acquired lesion vascular images according to the expansion sampling priority of all the lesion vascular images, until the expanded lesion vascular images reach a preset first number, and taking the expanded lesion vascular images and the normal vascular images as the expanded training model.

10. A peripheral vascular image-assisted segmentation system, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of a peripheral vascular image-assisted segmentation method as described in any one of claims 1 to 9 are implemented.

Images