Tumor medical image processing method and system for tumor ablation treatment system

Abstract

The present invention relates to the field of regional segmentation technology, and specifically to a tumor medical image processing method and system for a tumor ablation treatment system. The method performs initial segmentation on the CT image of the diseased part to obtain the tissue region, screens out the initial abnormal region, and screens the final abnormal region based on the difference in the degree of shape change between the initial abnormal region and its analysis region, and the difference in the number of images between the CT image of the diseased part where the initial abnormal region and its analysis region are located, and then screens the mis-segmented region; the gradient value of the edge pixel points of the mis-segmented region is adjusted based on the difference in the position distribution of the edge pixel points of the mis-segmented region and its analysis region, and then the adjusted gradient value is used to re-segment the CT image of the diseased part. The present invention effectively improves the accuracy of regional segmentation of the CT image of the diseased part.

Description

Technical Field

[0001] The present invention relates to the technical field of region segmentation, and in particular to a tumor medical image processing method and system for a tumor ablation treatment system. Background Art

[0002] Ablation therapy is a minimally invasive tumor treatment method that has developed rapidly in recent years. Because different tissue sites have high structural complexity and functional diversity, the boundaries of various tissues are often blurred in computed tomography (CT) images, especially the transition area between tumors and normal tissues. This complexity makes it difficult to determine the boundary between tumors and surrounding tissues. Therefore, when using a tumor ablation therapy system to treat tumors in patients, it is very important to accurately segment the tumor and normal tissue.

[0003] Existing methods usually use regional segmentation algorithms to segment normal tissue and tumor areas in tumor CT images. However, due to the blurred boundaries of tumor tissue and the possible presence of artifacts in CT images, the segmentation effect of tumor and surrounding tissue in CT images is poor, making the regional segmentation of CT images inaccurate. Summary of the invention

[0004] In order to solve the technical problem that the blurred boundaries and artifacts of tumor tissues lead to inaccurate regional segmentation of CT images by the regional segmentation algorithm, the purpose of the present invention is to provide a tumor medical image processing method and system for a tumor ablation treatment system. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present invention provides a tumor medical image processing method for a tumor ablation treatment system, the method comprising:

[0006] Acquire CT images of several layers of diseased parts of the patient;

[0007] Performing regional segmentation on the CT image of the diseased part to obtain tissue regions, and screening the initial abnormal region in the CT image of the diseased part from the tissue regions; obtaining the analysis region of each initial abnormal region, and screening the final abnormal region from the initial abnormal region according to the difference in the degree of shape change between each initial abnormal region and its analysis region, and the difference in the number of images between each initial abnormal region and the CT image of the diseased part where the analysis region is located;

[0008] According to the difference in edge fluctuation degree and shape between the final abnormal region and the analysis region, the mis-segmented region is screened out from the final abnormal region;

[0009] According to the difference in the position distribution of edge pixels of each mis-segmented area and its analysis area, the gradient value of the edge pixels of each mis-segmented area is adjusted, and the CT image of the diseased part is re-segmented using the adjusted gradient value.

[0010] Furthermore, the screening of the initial abnormal area in the CT image of the diseased part from the tissue area includes:

[0011] For each layer of the CT image of the diseased part, an adjacent region of each tissue region in the CT image of the diseased part is obtained, where edge pixel points on the edge of the adjacent region are edge pixel points on the edge of each tissue region;

[0012] The mean of the grayscale values ​​of all pixels in each tissue region is taken as the grayscale concentration value of each tissue region; the cumulative sum of the absolute values ​​of the differences between each tissue region and the grayscale concentration values ​​of all its adjacent regions is calculated, and recorded as the grayscale specific index of each tissue region;

[0013] Obtain the arithmetic mean difference of the grayscale values ​​of the pixels in each tissue area;

[0014] According to the grayscale specific index and the arithmetic mean difference, an abnormal index of each tissue region is obtained; the grayscale specific index and the arithmetic mean difference are both positively correlated with the abnormal index;

[0015] The tissue region with a value greater than the first preset threshold is taken as the initial abnormal region in the CT image of the diseased part.

[0016] Furthermore, the screening of the final abnormal area from the initial abnormal area includes:

[0017] For each initial abnormal region, the total number of pixels in the initial abnormal region is recorded as the area index of the corresponding region, and the total number of edge pixels on the edge of the initial abnormal region is recorded as the edge index of the corresponding region; according to the difference between the area index of each initial abnormal region and the analysis region, the difference between the edge index and the abnormal index of each initial abnormal region, the image change index of each initial abnormal region is obtained;

[0018] According to the difference in the number of images between each initial abnormal area and the CT image of the diseased part where the analysis area is located and the difference in the image change index, the abnormal change index of each initial abnormal area is obtained; and the initial abnormal area greater than the second preset threshold is used as the final abnormal area in the CT image of the diseased part.

[0019] Furthermore, the step of obtaining an abnormal change index of each initial abnormal area includes:

[0020] For each initial abnormal region, the number of images between the CT images of the diseased part of the initial abnormal region and each of its analysis regions is recorded as the interval image index between the initial abnormal region and each of its analysis regions; the absolute value of the difference between the image change index of the initial abnormal region and each of its analysis regions is calculated, and recorded as the interval change index between the initial abnormal region and each of its analysis regions;

[0021] The product of the absolute value of the difference between the interval change index of the initial abnormal region and all its analysis regions and the absolute value of the difference between the interval image index is normalized to obtain the abnormal change index of each initial abnormal region.

[0022] Furthermore, the step of screening out mis-segmented regions from the final abnormal regions includes:

[0023] For each final abnormal region, obtain the chain code value of each edge pixel on the edge of the final abnormal region; calculate the average of the absolute values ​​of the differences between each edge pixel on the edge of the final abnormal region and the chain code values ​​of its adjacent edge pixels, and record them as the local fluctuation index of each edge pixel on the edge of the final abnormal region; take the average of the local fluctuation indexes of all edge pixels on the edge of the final abnormal region as the edge fluctuation index of the final abnormal region;

[0024] Selecting any one of the analysis areas of the final abnormal area as an adjacent interval area of ​​the final abnormal area, and obtaining a shape edge difference index of the final abnormal area according to a difference between the edge fluctuation index of the final abnormal area and the adjacent interval area and a difference between the abnormal indexes;

[0025] The average value of the shape edge difference index of all final abnormal regions in each abnormal region sequence is used as the judgment threshold of the corresponding abnormal region; if the shape edge difference index of the final abnormal region is greater than the judgment threshold of the abnormal region sequence where the final abnormal region is located, the final abnormal region is used as the mis-segmented region.

[0026] Furthermore, the method of adjusting the gradient value of the edge pixel points of each mis-segmented area according to the position distribution difference between the edge pixel points of each mis-segmented area and its analysis area, and re-segmenting the CT image of the diseased part by using the adjusted gradient value, includes:

[0027] According to the difference in the position distribution of edge pixels of each mis-segmented area and its analysis area, possible abnormal edge indicators of each edge pixel on the edge of each mis-segmented area are obtained;

[0028] The sum of the abnormal edge possible index and the constant 1 is used as the adjustment coefficient of each edge pixel point on the edge of each mis-segmented area;

[0029] For each layer of the CT image of the diseased part, edge detection is performed on the CT image of the diseased part to obtain the gradient value of the edge pixel point on the edge of the mis-segmented area in the CT image of the diseased part, and the gradient value is weighted by the adjustment coefficient to obtain the optimized gradient value of the edge pixel point on the edge of the mis-segmented area;

[0030] Based on the optimized gradient value, the CT image of the diseased part is re-segmented using a watershed algorithm.

[0031] Furthermore, the step of obtaining possible abnormal edge indicators of each edge pixel point on the edge of each mis-segmented area includes:

[0032] For each final abnormal region, the distance between each edge pixel point on the edge of the final abnormal region and its centroid is used as the analysis distance of the corresponding edge pixel point;

[0033] Determine a starting point on the edge of the final abnormal region, and start from the starting point, sequentially arrange edge pixel points on the edge of the final abnormal region to obtain an edge point sequence of the final abnormal region;

[0034] For each mis-segmented region, the DTW algorithm is used to match the edge point sequence of the mis-segmented region and each of its analysis regions, and the average of the analysis distances of all edge pixel points that match each edge pixel point in the edge point sequence of the analysis region is used as the comparison distance of each edge pixel point in the edge point sequence of the mis-segmented region;

[0035] The cumulative sum of the absolute values ​​of the differences between the analysis distance and the control distance of each edge pixel point on the edge of the mis-segmented area is calculated, and the cumulative sum is negatively correlated and normalized to obtain a possible abnormal edge indicator of the corresponding edge pixel point.

[0036] Further, obtaining the analysis area of ​​each initial abnormal area includes:

[0037] The initial abnormal regions corresponding to each other in the CT images of all diseased parts are arranged in sequence to obtain a plurality of abnormal region sequences; for each abnormal region sequence, the initial abnormal region adjacent to each initial abnormal region in the abnormal region sequence is used as the analysis region of each initial abnormal region.

[0038] Furthermore, the method for performing regional segmentation on the CT image of the diseased part is a watershed algorithm.

[0039] In a second aspect, another embodiment of the present invention provides a tumor medical image processing system for a tumor ablation treatment system, the system comprising:

[0040] A data acquisition module, used to obtain CT images of several layers of diseased parts of the patient;

[0041] The abnormal region screening module is used to perform regional segmentation on the CT image of the diseased part to obtain tissue regions, and screen the initial abnormal region in the CT image of the diseased part from the tissue regions; obtain the analysis region of each initial abnormal region, and screen the final abnormal region from the initial abnormal region according to the difference in the degree of shape change between each initial abnormal region and its analysis region, and the difference in the number of images between each initial abnormal region and the CT image of the diseased part where the analysis region is located;

[0042] A mis-segmentation region screening module is used to screen mis-segmentation regions from the final abnormal region based on differences in edge fluctuations and shapes between the final abnormal region and its analysis region;

[0043] The regional segmentation optimization module is used to adjust the gradient value of the edge pixel points of each mis-segmented area according to the position distribution difference of the edge pixel points of each mis-segmented area and its analysis area, and use the adjusted gradient value to re-segment the CT image of the diseased part.

[0044] The present invention has the following beneficial effects:

[0045] In an embodiment of the present invention, an initial abnormal region is selected from a tissue region obtained by the initial segmentation of a CT image of a diseased part, and the initial abnormal region represents a tumor region. Due to the patient's breathing or heartbeat during scanning, artifacts may appear in the image, resulting in the destruction of regional continuity of the tissue structure in adjacent layer CT images, causing the tissue structure to appear discontinuous in different layer CT images, but the tissue structure of the continuous layer still presents similar features as a whole and shows regular changes with the patient's breathing and heartbeat. The difference in the number of images between the initial abnormal region and the CT image of the diseased part where the analysis region is located reflects regular change information, and the difference in the degree of shape change between the initial abnormal region and the analysis region reflects the structural similarity feature. The above factors are combined to select the final abnormal region, which is a region where the tumor region changes abnormally due to artifacts formed by breathing or heartbeat. Under normal circumstances, the edge and shape of the region are continuous in the multi-layer CT image. According to the difference in the edge fluctuation degree and shape difference between the final abnormal region and the analysis region, the mis-segmented region with abnormal change is determined, and the mis-segmented region causes the deviation of regional segmentation. The positions of tumor areas in continuous layers are relatively close. Under normal circumstances, the positions of the mis-segmented areas and their analysis areas are relatively close. According to the position segmentation difference of the edge pixels of the mis-segmented areas and their analysis areas, the gradient information of the mis-segmented areas is adjusted, and the CT images of the diseased parts are re-segmented to make the edges of the tumor areas and the normal tissue areas clearer and more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] 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.

[0047] Figure 1 A flowchart of the steps of a tumor medical image processing method for a tumor ablation treatment system provided by one embodiment of the present invention;

[0048] Figure 2 A flowchart of the steps of a final tissue region screening method provided by one embodiment of the present invention;

[0049] Figure 3 A system structure diagram of a tumor medical image processing system for a tumor ablation treatment system provided by one embodiment of the present invention;

[0050] Figure 4 A schematic diagram of a computer device of a tumor medical image processing device of a tumor ablation treatment system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] 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 tumor medical image processing method and system of a tumor ablation treatment system proposed by the present invention, its specific implementation, structure, characteristics 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.

[0052] 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.

[0053] The following describes in detail a tumor medical image processing method and a specific scheme of the system for tumor ablation treatment system provided by the present invention in conjunction with the accompanying drawings.

[0054] Embodiment 1:

[0055] The present invention proposes a tumor medical image processing method for a tumor ablation treatment system, please refer to Figure 1 , which shows a flowchart of the steps of a tumor medical image processing method of a tumor ablation treatment system provided by an embodiment of the present invention, the method comprising:

[0056] Step S1: Acquire several layers of CT images of the patient's diseased part.

[0057] Specifically, an X-ray tomography scanner is used to scan the patient's diseased part to obtain a multi-layer initial diseased part CT image; in order to reduce the impact of noise on image analysis, the initial patient part CT image is denoised to obtain a patient part CT image.

[0058] It should be noted that the embodiment of the present invention uses Gaussian filtering for denoising, and the specific method is not introduced here, which is a technical means well known to those skilled in the art; CT images are usually grayscale images.

[0059] Step S2: performing regional segmentation on the CT image of the diseased part to obtain tissue regions, and screening the initial abnormal regions in the CT image of the diseased part from the tissue regions; obtaining the analysis region of each initial abnormal region, and screening the final abnormal region from the initial abnormal region according to the difference in the degree of shape change between each initial abnormal region and its analysis region, and the difference in the number of images between each initial abnormal region and the CT image of the diseased part where the analysis region is located.

[0060] The human body has a complex tissue structure, and different parts have different textures and shapes. In order to improve the accuracy of tumor feature analysis, the embodiment of the present invention uses a watershed algorithm to perform regional segmentation on the CT image of the diseased part, and preliminarily divides the CT image of the diseased part into different tissue regions. The watershed algorithm is a well-known technology for those skilled in the art and will not be described in detail here.

[0061] The diseased part presents different characteristics in different layers of CT images. Since the CT image scanning process affects the characteristics of the tissue area, the initial segmentation of the tissue area may have unclear or fuzzy boundaries. This unclear area division will make it impossible for doctors to accurately locate the specific location and range of the tumor, thus affecting the diagnosis and subsequent treatment. Therefore, it is necessary to analyze the grayscale characteristics of the tissue area and obtain abnormal indicators to screen the initial abnormal area.

[0062] Preferably, in some possible implementation modes of the embodiments of the present invention, the method for obtaining the abnormal index includes: for each layer of the CT image of the diseased part, obtaining the adjacent areas of each tissue area in the CT image of the diseased part, and the edge pixels on the edge of the adjacent areas are the edge pixels on the edge of each tissue area; taking the mean of the grayscale values ​​of all pixels in each tissue area as the grayscale concentration value of each tissue area; calculating the cumulative sum of the absolute values ​​of the differences between the grayscale concentration values ​​of each tissue area and all its adjacent areas, and recording it as the grayscale specific index of each tissue area; obtaining the arithmetic mean difference of the grayscale values ​​of the pixels in each tissue area; obtaining the abnormal index of each tissue area according to the grayscale specific index and the arithmetic mean difference; the grayscale specific index and the arithmetic mean difference are positively correlated with the abnormal index.

[0063] Because the boundaries of malignant tumors are usually unclear, the tumor area is the focus of analysis. In order to conduct targeted analysis of tumors, it is necessary to first identify the tumor area from the tissue area. The tumor may contain different components, such as necrosis, hemorrhage, or fluid accumulation, which makes the grayscale of the tumor area uneven, while the grayscale of the normal tissue area is relatively uniform; the grayscale characteristics of the tumor area are more obvious than those of the surrounding normal tissue area, that is, the grayscale of the tumor area is significantly higher or lower than the grayscale of the surrounding normal tissue area.

[0064] The adjacent area of ​​the tissue area is the tissue area around the tissue area. The grayscale concentration value reflects the overall grayscale level of the tissue area. The grayscale feature prominence of the tissue area is measured by the accumulation and sum of the absolute values ​​of the grayscale concentration values ​​of the tissue area and all its adjacent areas, and the grayscale specific index is obtained. If the grayscale specific index is larger, it means that the grayscale feature of the tissue area is more prominent than that of the surrounding tissue area, then the abnormality of the tissue area is greater, and the possibility that the tissue area is a tumor tissue is greater. The arithmetic mean difference of the grayscale values ​​of the pixels in the tissue area reflects the grayscale uniformity of the tissue area. If the arithmetic mean difference is larger, it means that the grayscale of the tissue area is more uneven, then the abnormality of the tissue area is greater, and the possibility that the tissue area is a tumor tissue is greater. Therefore, the arithmetic mean difference and the grayscale specific index are positively correlated with the abnormality index.

[0065] In a specific implementation of the embodiment of the present invention, the abnormal index of the tissue region is expressed by the formula:

[0066]

[0067] Where G is the abnormality index of each tissue area; is the grayscale concentration value of each tissue area; T is the total number of adjacent areas of each tissue area; is the grayscale concentration value of the tth adjacent area of ​​each tissue area; is the grayscale specific index of each tissue area; U is the total number of pixels in each tissue area; is the gray value of the u-th pixel in each tissue area; It is the arithmetic mean difference of the gray value of pixels in each tissue area; is the absolute value function; Norm is the normalization function.

[0068] The greater the abnormal index of the tissue region, the greater the possibility that the tissue region is a tumor region. Therefore, the tissue region greater than the first preset threshold is used as the initial abnormal region in the CT image of the diseased part; the initial abnormal region represents the tumor tissue in the CT image of the diseased part. It should be noted that in the embodiment of the present invention, the first preset threshold is an empirical value of 0.8, and the implementer can set it according to the specific situation.

[0069] The known tumor structure will be divided into multiple layers of CT images of the diseased part, and each layer of CT images of the diseased part can show the anatomical structure and pathological state of the tumor structure at a specific depth. Since the CT images of the diseased part are obtained by scanning the body layer by layer, the tissue structure has regional continuity in adjacent layers of CT images; however, during the scanning process, the patient's breathing, heartbeat or other unconscious movements will cause artifacts in the CT image, causing a large degree of change in some areas, resulting in discontinuity of the tissue structure in different layers of CT images, but the tissue structure of adjacent layers still presents similar characteristics as a whole. Therefore, the analysis area of ​​the initial abnormal area is obtained, and the initial abnormal area and its analysis area can be regarded as a continuous layer of tumor structure.

[0070] In this embodiment, the initial abnormal regions corresponding to each other in the CT images of all diseased parts are arranged in sequence to obtain a plurality of abnormal region sequences; for each abnormal region sequence, in the abnormal region sequence, the initial abnormal region adjacent to each initial abnormal region is used as the analysis region of each initial abnormal region.

[0071] It should be noted that each layer of the CT image of the diseased part has a corresponding scanning time, and arranging all the CT images of the diseased part in the order of scanning time can present a three-dimensional model of the diseased part. The initial abnormal areas corresponding to each other in all the CT images of the diseased part are arranged in the order of scanning time of the CT images of the diseased part where the initial abnormal areas are located, to obtain several abnormal area sequences; the initial abnormal areas in each abnormal area sequence can constitute a tumor structure. In the abnormal area sequence, for the remaining initial abnormal areas except the first and last initial abnormal areas, the adjacent previous and adjacent next initial abnormal areas of each initial abnormal area are used as the analysis areas of each initial abnormal area; the initial abnormal area has two analysis areas. It should be noted that in this embodiment, the first and last initial abnormal areas in the abnormal area sequence are directly used as the final abnormal areas.

[0072] In other embodiments, the centroid of the initial abnormal area in the CT image of the diseased part is obtained, and a layer of the CT image of the diseased part is randomly recorded as the target CT image, and an initial abnormal area is randomly selected from the target CT image as the target area, and the pixel point at the same position of the centroid of the target area in the remaining CT images of the diseased part except the target CT image is used as the reference point of the target area; for each CT image of the diseased part except the target CT image, in the CT image of the diseased part, it is determined whether there is a centroid of the initial abnormal area within a preset range of the reference point of the target area, and if so, the initial abnormal area corresponding to the centroid closest to the reference point of the target area is used as the initial abnormal area corresponding to the target area in the CT image of the diseased part; if not, the target area does not have a corresponding initial abnormal area in the CT image of the diseased part; the target area and its corresponding initial abnormal area in the remaining CT images of the diseased part except the target CT image are arranged in the order of the scanning time of the CT image of the diseased part where the area is located, and an abnormal area sequence is obtained. According to the above method, all abnormal area sequences in all CT images of the diseased part are obtained.

[0073] It should be noted that, since the multi-layer CT images of the diseased part are obtained by scanning the body layer by layer, the image changes between adjacent layer CT images are usually small, and the positions of the corresponding areas of the same tumor structure in different diseased part CT images are basically the same; if the centroids of these areas are mapped to the same plane, these centroids are relatively concentrated. The preset range of the reference point in this embodiment is: an area formed by taking the reference point as the center and 50 as the radius, and the radius can be set according to the actual situation.

[0074] Although the patient's breathing and heartbeat may cause the CT images of the tissue structures of the continuous layers to be discontinuous, the tissue structures of the continuous layers still show similar characteristics as a whole and change regularly with the patient's breathing and heartbeat. The difference in the number of images between the initial abnormal region and the CT images of the diseased parts of the two analysis regions reflects the regular change information, and the difference in the degree of shape change between the initial abnormal region and its analysis region reflects the structural similarity characteristics. After comprehensive analysis of the above factors, the final abnormal region is selected from the initial abnormal region, which is the region where the tumor region changes abnormally due to artifacts formed by breathing or heartbeat.

[0075] See also Figure 2 , which shows a flow chart of the steps of a final abnormal area screening method provided by an embodiment of the present invention, the method comprising:

[0076] Step S210: For each initial abnormal region, the total number of pixels in the initial abnormal region is recorded as the area index of the corresponding region, and the total number of edge pixels on the edge of the initial abnormal region is recorded as the edge index of the corresponding region; according to the difference in area index of each initial abnormal region and its analysis region, the difference in edge index and the abnormal index of each initial abnormal region, the image change index of each initial abnormal region is obtained.

[0077] As the scanning direction changes, the CT image will gradually change, and the degree of change of the multi-layer tissue area divided by the tissue structure is basically the same; the image changes of the same tissue structure in the CT images of the diseased parts of different layers are analyzed by the area and edge length of the region. The area index reflects the area information of the initial abnormal area, and the edge index reflects the edge length information of the initial abnormal area; if the difference in the area index and the difference in the edge index between the initial abnormal area and its analysis area are greater, it means that the degree of change between the initial abnormal area and its analysis area is greater, and the image change index is greater.

[0078] Since the growth and development of a tumor may lead to rapid changes in its morphology, boundaries, and internal structure, the larger the abnormal index of the initial abnormal region, the greater the possibility that the initial abnormal region represents actual tumor tissue, and the greater the possibility that the initial abnormal region will change, the greater the image change index.

[0079] Therefore, the difference between the area index of the initial abnormal region and its analysis region, the difference between the edge index and the abnormality index of the initial abnormal region are all positively correlated with the image change index. In an embodiment of the present invention, the product of the absolute value of the difference between the area index of each initial abnormal region and its analysis region, the absolute value of the difference between the edge index and the abnormality index of each initial abnormal region is used as the image change index of each initial abnormal region.

[0080] In the embodiment of the present invention, other basic mathematical operations can also be used to construct the absolute value of the difference between the area index of the initial abnormal region and its analysis region, the absolute value of the difference between the edge index and the correlation between each initial abnormal index and the image change index, which will not be limited or elaborated here.

[0081] Step S220: Obtain the abnormal change index of each initial abnormal area according to the difference in the number of images and the difference in the image change index between each initial abnormal area and the CT image of the diseased part where its analysis area is located; and take the initial abnormal area greater than the second preset threshold as the final abnormal area in the CT image of the diseased part.

[0082] Preferably, in some possible implementation modes of the embodiments of the present invention, the method for obtaining the abnormal change index includes: for each initial abnormal area, recording the number of images between the CT images of the diseased part of the initial abnormal area and each of its analysis areas as the interval image index of the initial abnormal area and each of its analysis areas; calculating the absolute value of the difference between the image change index of the initial abnormal area and each of its analysis areas, recorded as the interval change index of the initial abnormal area and each of its analysis areas; normalizing the product of the absolute value of the difference between the interval change index of the initial abnormal area and all its analysis areas and the absolute value of the difference between the interval image index to obtain the abnormal change index of each initial abnormal area.

[0083] As the scanning direction changes, the state of the CT image gradually changes, and the degree of change of the multi-layer tissue area divided by the tissue structure is basically consistent, and these degrees of change are usually small; considering the impact of motion artifacts on the analysis, because breathing and heartbeat have regularity, as the scanning proceeds, the CT images of the tissue structures of the continuous layers show regular changes, that is, the number of images between the CT images of the tissue structures of the continuous layers is relatively close. Therefore, if the number of images between the CT images of the tissue structures of the continuous layers is relatively close and the degree of change of the tissue structures of the continuous layers corresponding to the initial abnormal areas is closer, the possibility of abnormal changes in the tumor area caused by artifacts formed by breathing or heartbeat in the initial abnormal areas corresponding to the tissue structures of these continuous layers is smaller. When the absolute value of the difference between the interval change index of the initial abnormal area and its analysis area and the absolute value of the difference between the interval image index are larger, the possibility of abnormal changes in the initial abnormal area is greater.

[0084] In a specific implementation of the embodiment of the present invention, the abnormal change index of the initial abnormal area is expressed by the formula:

[0085]

[0086] Where, B is the abnormal change index of each initial abnormal area; P is the image change index of each initial abnormal area; It is the image change index of the first analysis area of ​​each initial abnormal area; The image change index of the second analysis area for each initial abnormal area; It is the interval change index between each initial abnormal area and its first analysis area; It is the interval change index between each initial abnormal area and its second analysis area; It is the interval image index between each initial abnormal area and its first analysis area; It is the interval image index between each initial abnormal area and its second analysis area; is the absolute value function; Norm is the normalization function.

[0087] If the abnormal change index of the initial abnormal region is larger, it means that the change regularity between the initial abnormal region and its analysis region is worse, and the possibility of abnormal change in the initial abnormal region is greater, then the initial abnormal region greater than the second preset threshold is used as the final abnormal region in the CT image of the diseased part. It should be noted that in this embodiment, the second preset threshold is an empirical value of 0.9, and the implementer can set it according to the specific situation.

[0088] Step S3: Filter mis-segmented areas from the final abnormal area according to differences in edge fluctuation and shape between the final abnormal area and its analysis area.

[0089] Under normal circumstances, the edge and shape of the region are continuous in different layers of CT images. If the regional change of a certain layer of regional images in the tissue structure of the continuous layer is significantly greater than that of other layers, the regional segmentation of the layer of regional images may have certain deviations due to various factors. Because the degree of abnormal change of the final abnormal region in adjacent images should be relatively close, the degree of abnormal change of the region is reflected by the edge fluctuation degree and shape of the region, so as to identify the mis-segmented region.

[0090] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the shape edge difference index includes: for each final abnormal area, obtaining the chain code value of each edge pixel on the edge of the final abnormal area; calculating the average of the absolute values ​​of the difference between each edge pixel on the edge of the final abnormal area and the chain code values ​​of its adjacent edge pixels, and recording it as the local fluctuation index of each edge pixel on the edge of the final abnormal area; taking the average of the local fluctuation indexes of all edge pixels on the edge of the final abnormal area as the edge fluctuation index of the final abnormal area; selecting any one analysis area from the analysis areas of the final abnormal area as the adjacent interval area of ​​the final abnormal area, and obtaining the shape edge difference index of the final abnormal area according to the difference in edge fluctuation index and the difference in abnormal index between the final abnormal area and its adjacent interval area. Among them, the method for obtaining the chain code value is a well-known technology and will not be repeated here.

[0091] The chain code value of the edge pixel of the final abnormal area reflects the change trend of the edge of the final abnormal area. The chain code value of each edge pixel on the edge of the final abnormal area represents the change direction from the current edge pixel to the next edge pixel in the clockwise direction. The local fluctuation index of each edge pixel on the edge of the final abnormal area shows the fluctuation degree of the local edge where the edge pixel is located; the mean value of the local fluctuation index of all edge pixels on the edge of the final abnormal area reflects the overall fluctuation characteristics of the edge of the final abnormal area. The abnormal index reflects the area characteristics and edge length characteristics of the final abnormal area, and the two together determine the shape characteristics of the final abnormal area.

[0092] In the embodiment of the present invention, in the abnormal region sequence, the next adjacent initial abnormal region of each final abnormal region is used as the adjacent interval region of the final abnormal region. If the difference in edge fluctuation index between the final abnormal region and its adjacent interval region and the difference in abnormal index are both greater, it means that the change in edge fluctuation and shape characteristics of the final abnormal region in the adjacent layer is more obvious, and the shape edge difference index is greater. Therefore, the difference in edge fluctuation index between the final abnormal region and its adjacent interval region and the difference in abnormal index and the shape edge difference index are positively correlated.

[0093] In the embodiment of the present invention, the product of the absolute value of the difference between the edge fluctuation index of the final abnormal region and its adjacent interval region and the absolute value of the difference between the abnormal indexes is used as the shape edge difference index of the final abnormal region. The larger the shape edge index, the greater the possibility that the final abnormal region is mis-segmented.

[0094] The average of the shape edge difference indicators of all final abnormal regions in each abnormal region sequence is used as the judgment threshold of the corresponding abnormal region; if the shape edge difference indicator of the final abnormal region is greater than the judgment threshold of the abnormal region sequence where the final abnormal region is located, the final abnormal region is regarded as the mis-segmented region.

[0095] Step S4: According to the position distribution difference of the edge pixels of each mis-segmented area and its analysis area, the gradient value of the edge pixels of each mis-segmented area is adjusted, and the CT image of the diseased part is re-segmented using the adjusted gradient value.

[0096] The positions of the tumor areas in the continuous layers are relatively close. By analyzing the position segmentation differences of the edge pixels of the mis-segmented area and its analysis area, possible indicators of abnormal edges are obtained to measure the possibility that the edge pixels belong to the actual tumor edge; the gradient values ​​of the edge pixels of the mis-segmented area are adjusted to improve the accuracy of regional segmentation.

[0097] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining possible indicators of abnormal edges includes:

[0098] For each final abnormal region, the distance between each edge pixel point on the edge of the final abnormal region and its centroid is used as the analysis distance of the corresponding edge pixel point; the starting point on the edge of the final abnormal region is determined, and the edge pixel points on the edge of the final abnormal region are arranged in sequence from the starting point to obtain the edge point sequence of the final abnormal region; for each mis-segmented region, the DTW algorithm is used to match the edge point sequence of the mis-segmented region and each of its analysis regions, and the average of the analysis distances of all edge pixel points that match each edge pixel point in the edge point sequence of the analysis region is used as the reference distance of each edge pixel point in the edge point sequence of the mis-segmented region; the cumulative sum of the absolute values ​​of the differences between the analysis distance of each edge pixel point on the edge of the mis-segmented region and its reference distance is calculated, and the cumulative sum is negatively correlated and normalized to obtain the possible abnormal edge indicator of the corresponding edge pixel point. Among them, the Dynamic Time Warping (DTW) algorithm is a well-known technology for those skilled in the art and will not be described in detail here.

[0099] The method for determining the starting point on the edge of the final abnormal area is as follows: for the analysis distance of all edge pixel points on the edge of the final abnormal area, select the edge pixel point corresponding to the largest analysis distance as the initial point; record the line segment connecting each initial point and the centroid of the final abnormal area as the screening line segment of each initial point; obtain the angle between the screening line segment of all initial points and the horizontal right direction, and use the initial point corresponding to the screening line segment corresponding to the smallest angle as the starting point on the edge of the final abnormal area.

[0100] In this embodiment, starting from the starting point on the edge of the final abnormal area, the edge pixels on the edge of the final abnormal area are arranged in sequence in a clockwise direction to obtain the edge point sequence of the final abnormal area. The difference in the analysis distance between each edge pixel in the edge point sequence of the mis-segmented area and the edge pixel matching it in the edge point sequence of the analysis area presents the position difference between each edge pixel in the mis-segmented area and the edge pixel matching it. If the difference is larger, it means that the mis-segmented area has a greater degree of deviation than its analysis area, and the edge pixel belongs to the actual tumor edge of the edge of the area. Therefore, the cumulative sum of the absolute values ​​of the difference between the analysis distance of each edge pixel on the edge of the mis-segmented area and its control distance is negatively correlated with the possible abnormal edge index.

[0101] It should be noted that each edge pixel has two reference distances; if an edge pixel in the mis-segmented area has no matching edge pixel in one or two analysis areas, that is, the edge pixel has no or only one reference distance, the abnormal edge possible index of the edge pixel in the mismatched area is directly set to a constant of 0.

[0102] This solution uses the watershed algorithm to segment the CT images of the diseased part. Usually, the grayscale image is converted into a gradient image for analysis. Therefore, it is necessary to use the abnormal edge possible index to adjust the gradient value of the edge pixel points of the mis-segmented area to improve the accuracy of the regional segmentation. The larger the abnormal edge possible index of the edge pixel point, the greater the possibility that the edge pixel point belongs to the edge of the tumor tissue. Because the edge of the tumor tissue is relatively fuzzy, the larger the gradient value, the clearer the edge of the mis-segmented area. By increasing the gradient value of the edge pixel point, the accuracy of the regional segmentation can be improved.

[0103] The specific method for adjusting the gradient value is: taking the sum of the possible index of the abnormal edge and the constant 1 as the adjustment coefficient of each edge pixel point on the edge of each mis-segmented area; for each layer of the CT image of the diseased part, performing edge detection on the CT image of the diseased part to obtain the gradient value of the edge pixel points on the edge of the mis-segmented area in the CT image of the diseased part, and using the adjustment coefficient to weight the gradient value to obtain the optimized gradient value of the edge pixel points on the edge of the mis-segmented area.

[0104] For each CT image of the diseased part, the optimized gradient values ​​of the edge pixels of the mis-segmented area in the CT image of the diseased part are obtained, and the gradient values ​​of the edge pixels of the non-mis-segmented area remain unchanged, so as to obtain the optimized gradient image of the CT image of the diseased part. The optimized gradient image is segmented by the watershed algorithm, so as to achieve more accurate regional segmentation of the CT image of the diseased part, so that the edge of the tumor area and the normal tissue area will be clearer and more accurate.

[0105] It should be noted that the present embodiment uses the Canny operator to perform edge detection on the CT image of the diseased part, wherein the Canny operator is a well-known technology to those skilled in the art and will not be described in detail here.

[0106] By observing the status of the same tissue structure in different CT images, doctors can more intuitively evaluate tumor characteristics and tumor growth trends, assist doctors in determining the location and range of the tumor structure, and develop more reasonable tumor ablation treatment plans and optimize treatment plans.

[0107] So far, the present invention is completed.

[0108] Embodiment 2:

[0109] The present invention proposes a tumor medical image processing system for a tumor ablation treatment system, please refer to Figure 3 , which shows a system structure diagram of a tumor medical image processing system of a tumor ablation treatment system provided by an embodiment of the present invention, the system comprising:

[0110] The data acquisition module 510 is used to obtain CT images of several layers of diseased parts of the patient;

[0111] The abnormal region screening module 520 is used to perform regional segmentation on the CT image of the diseased part to obtain tissue regions, and screen the initial abnormal region in the CT image of the diseased part from the tissue regions; obtain the analysis region of each initial abnormal region, and screen the final abnormal region from the initial abnormal region according to the difference in the degree of shape change between each initial abnormal region and its analysis region, and the difference in the number of images between each initial abnormal region and the CT image of the diseased part where the analysis region is located;

[0112] The mis-segmentation region screening module 530 is used to screen the mis-segmentation region from the final abnormal region according to the difference in edge fluctuation degree and shape between the final abnormal region and the analysis region;

[0113] The region segmentation optimization module 540 is used to adjust the gradient value of the edge pixel points of each mis-segmented region according to the position distribution difference of the edge pixel points of each mis-segmented region and its analysis region, and use the adjusted gradient value to re-segment the region of the CT image of the diseased part.

[0114] It should be noted that the device provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the tumor medical image processing system of a tumor ablation treatment system provided in the above embodiment and the tumor medical image processing method embodiment of a tumor ablation treatment system belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0115] Embodiment 3:

[0116] Figure 4 A computer device schematic diagram of a tumor medical image processing device of a tumor ablation treatment system provided by an embodiment of the present invention. Figure 4 As shown, the computer device includes: a memory 601, a processor 602, and a computer program 603 stored in the memory 601 and running on the processor 602, wherein when the processor 602 executes the computer program 603, the computer device can execute any tumor medical image processing method of the tumor ablation treatment system introduced above.

[0117] In addition, an embodiment of the present application also protects a device, which may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to execute a tumor medical image processing method for a tumor ablation treatment system provided in an embodiment of the present application.

[0118] In this embodiment, the functional modules of the device can be divided according to the above method example. For example, each functional module can be corresponded, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0119] It should be understood that the device provided in this embodiment is used to execute the above-mentioned tumor medical image processing method of a tumor ablation treatment system, and thus can achieve the same effect as the above-mentioned implementation method.

[0120] In the case of an integrated unit, the device may include a processing module and a storage module. When the device is applied to a device, the processing module may be used to control and manage the actions of the device. The storage module may be used to support the device to execute mutual program codes, etc.

[0121] The processing module may be a processor or a controller, which may implement or execute various exemplary logic blocks, modules and circuits included in the disclosure of this application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module may be a memory.

[0122] It should be noted that the sequence of the above embodiments of the present invention is only for description 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.

[0123] 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.

[0124] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A tumor medical image processing method for a tumor ablation treatment system, characterized in that: The method includes: Acquire CT images of several layers of diseased parts of the patient; Performing regional segmentation on the CT image of the diseased part to obtain tissue regions, and screening the initial abnormal region in the CT image of the diseased part from the tissue regions; obtaining the analysis region of each initial abnormal region, and screening the final abnormal region from the initial abnormal region according to the difference in the degree of shape change between each initial abnormal region and its analysis region, and the difference in the number of images between each initial abnormal region and the CT image of the diseased part where the analysis region is located; According to the difference in edge fluctuation degree and shape between the final abnormal region and the analysis region, the mis-segmented region is screened out from the final abnormal region; According to the difference in the position distribution of the edge pixels of each mis-segmented area and its analysis area, the gradient value of the edge pixels of each mis-segmented area is adjusted, and the CT image of the diseased part is re-segmented using the adjusted gradient value; The step of screening the initial abnormal region in the CT image of the diseased part from the tissue region comprises: For each layer of the CT image of the diseased part, an adjacent region of each tissue region in the CT image of the diseased part is obtained, where edge pixel points on the edge of the adjacent region are edge pixel points on the edge of each tissue region; The mean of the grayscale values ​​of all pixels in each tissue region is taken as the grayscale concentration value of each tissue region; the cumulative sum of the absolute values ​​of the differences between each tissue region and the grayscale concentration values ​​of all its adjacent regions is calculated, and recorded as the grayscale specific index of each tissue region; Obtain the arithmetic mean difference of the grayscale values ​​of the pixels in each tissue area; According to the grayscale specific index and the arithmetic mean difference, an abnormal index of each tissue region is obtained; the grayscale specific index and the arithmetic mean difference are both positively correlated with the abnormal index; Taking the tissue area larger than the first preset threshold as the initial abnormal area in the CT image of the diseased part; The screening of the final abnormal area from the initial abnormal area comprises: For each initial abnormal region, the total number of pixels in the initial abnormal region is recorded as the area index of the corresponding region, and the total number of edge pixels on the edge of the initial abnormal region is recorded as the edge index of the corresponding region; according to the difference between the area index of each initial abnormal region and the analysis region, the difference between the edge index and the abnormal index of each initial abnormal region, the image change index of each initial abnormal region is obtained; Obtaining the abnormal change index of each initial abnormal region according to the difference in the number of images between each initial abnormal region and the CT image of the diseased part where the analysis region is located and the difference in the image change index; taking the initial abnormal region greater than the second preset threshold as the final abnormal region in the CT image of the diseased part; The step of screening out mis-segmented regions from the final abnormal regions comprises: For each final abnormal region, obtain the chain code value of each edge pixel on the edge of the final abnormal region; calculate the average of the absolute values ​​of the differences between each edge pixel on the edge of the final abnormal region and the chain code values ​​of its adjacent edge pixels, and record them as the local fluctuation index of each edge pixel on the edge of the final abnormal region; take the average of the local fluctuation indexes of all edge pixels on the edge of the final abnormal region as the edge fluctuation index of the final abnormal region; Selecting any one of the analysis areas of the final abnormal area as an adjacent interval area of ​​the final abnormal area, and obtaining a shape edge difference index of the final abnormal area according to a difference between the edge fluctuation index of the final abnormal area and the adjacent interval area and a difference between the abnormal indexes; The average value of the shape edge difference index of all final abnormal regions in each abnormal region sequence is used as the judgment threshold of the corresponding abnormal region; if the shape edge difference index of the final abnormal region is greater than the judgment threshold of the abnormal region sequence where the final abnormal region is located, the final abnormal region is used as the mis-segmented region.

2. The tumor medical image processing method of a tumor ablation treatment system according to claim 1, characterized in that: The obtaining of the abnormal change index of each initial abnormal area includes: For each initial abnormal region, the number of images between the CT images of the diseased part of the initial abnormal region and each of its analysis regions is recorded as the interval image index between the initial abnormal region and each of its analysis regions; the absolute value of the difference between the image change index of the initial abnormal region and each of its analysis regions is calculated, and recorded as the interval change index between the initial abnormal region and each of its analysis regions; The product of the absolute value of the difference between the interval change index of the initial abnormal region and all its analysis regions and the absolute value of the difference between the interval image index is normalized to obtain the abnormal change index of each initial abnormal region.

3. The tumor medical image processing method of a tumor ablation treatment system according to claim 1, characterized in that: The method of adjusting the gradient value of the edge pixel points of each mis-segmented area according to the position distribution difference of the edge pixel points of each mis-segmented area and the analysis area, and re-segmenting the CT image of the diseased part by using the adjusted gradient value, comprises: According to the difference in the position distribution of edge pixels of each mis-segmented area and its analysis area, possible abnormal edge indicators of each edge pixel on the edge of each mis-segmented area are obtained; The sum of the abnormal edge possible index and the constant 1 is used as the adjustment coefficient of each edge pixel point on the edge of each mis-segmented area; For each layer of the CT image of the diseased part, edge detection is performed on the CT image of the diseased part to obtain the gradient value of the edge pixel point on the edge of the mis-segmented area in the CT image of the diseased part, and the gradient value is weighted by the adjustment coefficient to obtain the optimized gradient value of the edge pixel point on the edge of the mis-segmented area; Based on the optimized gradient value, the CT image of the diseased part is re-segmented using a watershed algorithm.

4. The tumor medical image processing method of a tumor ablation treatment system according to claim 3, characterized in that: The method of obtaining possible abnormal edge indicators of each edge pixel point on the edge of each mis-segmented area includes: For each final abnormal region, the distance between each edge pixel point on the edge of the final abnormal region and its centroid is used as the analysis distance of the corresponding edge pixel point; Determine a starting point on the edge of the final abnormal region, and starting from the starting point, sequentially arrange edge pixel points on the edge of the final abnormal region to obtain an edge point sequence of the final abnormal region; For each mis-segmented region, the DTW algorithm is used to match the edge point sequence of the mis-segmented region and each of its analysis regions, and the average of the analysis distances of all edge pixel points that match each edge pixel point in the edge point sequence of the analysis region is used as the comparison distance of each edge pixel point in the edge point sequence of the mis-segmented region; The cumulative sum of the absolute values ​​of the differences between the analysis distance and the control distance of each edge pixel point on the edge of the mis-segmented area is calculated, and the cumulative sum is negatively correlated and normalized to obtain a possible abnormal edge indicator of the corresponding edge pixel point.

5. The tumor medical image processing method of a tumor ablation treatment system according to claim 1, characterized in that: The step of obtaining the analysis area of ​​each initial abnormal area includes: The initial abnormal regions corresponding to each other in the CT images of all diseased parts are arranged in sequence to obtain a plurality of abnormal region sequences; for each abnormal region sequence, the initial abnormal region adjacent to each initial abnormal region in the abnormal region sequence is used as the analysis region of each initial abnormal region.

6. The tumor medical image processing method of a tumor ablation treatment system according to claim 1, characterized in that: The method for performing regional segmentation on the CT image of the diseased part is a watershed algorithm.

7. A tumor medical image processing system for a tumor ablation treatment system, characterized in that: The system includes: A data acquisition module, used to obtain CT images of several layers of diseased parts of the patient; The abnormal region screening module is used to perform regional segmentation on the CT image of the diseased part to obtain tissue regions, and screen the initial abnormal region in the CT image of the diseased part from the tissue regions; obtain the analysis region of each initial abnormal region, and screen the final abnormal region from the initial abnormal region according to the difference in the degree of shape change between each initial abnormal region and its analysis region, and the difference in the number of images between each initial abnormal region and the CT image of the diseased part where the analysis region is located; A mis-segmentation region screening module is used to screen mis-segmentation regions from the final abnormal region based on differences in edge fluctuations and shapes between the final abnormal region and its analysis region; A region segmentation optimization module is used to adjust the gradient value of the edge pixel points of each mis-segmented region according to the position distribution difference of the edge pixel points of each mis-segmented region and its analysis region, and re-segment the CT image of the diseased part using the adjusted gradient value; The step of screening the initial abnormal region in the CT image of the diseased part from the tissue region comprises: For each layer of the CT image of the diseased part, an adjacent region of each tissue region in the CT image of the diseased part is obtained, where edge pixel points on the edge of the adjacent region are edge pixel points on the edge of each tissue region; The mean of the grayscale values ​​of all pixels in each tissue region is taken as the grayscale concentration value of each tissue region; the cumulative sum of the absolute values ​​of the differences between each tissue region and the grayscale concentration values ​​of all its adjacent regions is calculated, and recorded as the grayscale specific index of each tissue region; Obtain the arithmetic mean difference of the grayscale values ​​of the pixels in each tissue area; According to the grayscale specific index and the arithmetic mean difference, an abnormal index of each tissue region is obtained; the grayscale specific index and the arithmetic mean difference are both positively correlated with the abnormal index; Taking the tissue area larger than the first preset threshold as the initial abnormal area in the CT image of the diseased part; The screening of the final abnormal area from the initial abnormal area comprises: For each initial abnormal region, the total number of pixels in the initial abnormal region is recorded as the area index of the corresponding region, and the total number of edge pixels on the edge of the initial abnormal region is recorded as the edge index of the corresponding region; according to the difference between the area index of each initial abnormal region and the analysis region, the difference between the edge index and the abnormal index of each initial abnormal region, the image change index of each initial abnormal region is obtained; Obtaining the abnormal change index of each initial abnormal region according to the difference in the number of images between each initial abnormal region and the CT image of the diseased part where the analysis region is located and the difference in the image change index; taking the initial abnormal region greater than the second preset threshold as the final abnormal region in the CT image of the diseased part; The step of screening out mis-segmented regions from the final abnormal regions comprises: For each final abnormal region, obtain the chain code value of each edge pixel on the edge of the final abnormal region; calculate the average of the absolute values ​​of the differences between each edge pixel on the edge of the final abnormal region and the chain code values ​​of its adjacent edge pixels, and record them as the local fluctuation index of each edge pixel on the edge of the final abnormal region; take the average of the local fluctuation indexes of all edge pixels on the edge of the final abnormal region as the edge fluctuation index of the final abnormal region; Selecting any one of the analysis areas of the final abnormal area as an adjacent interval area of ​​the final abnormal area, and obtaining a shape edge difference index of the final abnormal area according to a difference between the edge fluctuation index of the final abnormal area and the adjacent interval area and a difference between the abnormal indexes; The average value of the shape edge difference index of all final abnormal regions in each abnormal region sequence is used as the judgment threshold of the corresponding abnormal region; if the shape edge difference index of the final abnormal region is greater than the judgment threshold of the abnormal region sequence where the final abnormal region is located, the final abnormal region is used as the mis-segmented region.

Images