A method and system for processing hepatobiliary CT images

Abstract

The present invention relates to a hepatobiliary CT image processing method and system, and relates to the field of image processing technology. The hepatobiliary CT image processing method acquires hepatobiliary CT images, establishes a three-dimensional model of hepatobiliary tissue, and performs feature recognition analysis to obtain a hepatobiliary tissue feature dataset; performs comprehensive analysis to obtain a liver tissue health index and a gallbladder tissue health index; and performs comprehensive analysis to obtain a comprehensive hepatobiliary tissue health index, which is then compared with a preset hepatobiliary tissue comprehensive health assessment interval and marked as healthy based on the judgment and analysis results. By constructing a complete three-dimensional hepatobiliary tissue model and extracting multidimensional features, the present invention significantly improves the accuracy and comprehensiveness of hepatobiliary health assessment. By performing nonlinear integration of the analyzed liver tissue health index and gallbladder tissue health index, the calculated comprehensive hepatobiliary tissue health index can comprehensively assess the overall health status of the hepatobiliary system.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method and system for processing hepatobiliary CT images. Background Art

[0002] In the field of modern medical image analysis and hepatobiliary health assessment, accurate analysis of hepatobiliary tissue is a critical and widely applicable research topic. Traditional hepatobiliary health assessment methods mainly rely on simple parameter analysis of single indicators such as volume and density. However, these methods usually have problems such as single indicators and insufficient sensitivity to complex disease characteristics. In many practical applications, methods based on multidimensional features and comprehensive analysis have become a more attractive option due to their comprehensiveness and accuracy.

[0003] Despite the advantages mentioned above, multidimensional feature-based assessment methods still face many challenges in practical applications. For example, traditional methods only focus on the characteristics of a single region of the liver or gallbladder, ignoring the coordinated changes of the hepatobiliary system as a whole under pathological conditions. In addition, many assessment methods rely on simple weighted models and fail to fully capture the nonlinear correlations between hepatobiliary tissue characteristics, which easily leads to poor performance on complex lesions and makes it difficult to meet the needs of comprehensive health assessment.

[0004] Traditional technologies also face limitations in feature extraction and analysis when processing three-dimensional models of hepatobiliary tissue. Conventional methods lack innovative analytical methods for deep features such as the liver's geometric complexity and metabolically active areas. At the same time, they insufficiently explore features such as the distribution of gallstones and the uniformity of gallbladder wall thickness, which can easily lead to evaluation results that are difficult to fully reflect the actual health status of the hepatobiliary tissue. In addition, when synthesizing the hepatobiliary health index, traditional technologies fail to effectively integrate multidimensional features, which can easily lead to deviations and affect the ability to support early diagnosis and precise treatment of the disease. Summary of the Invention

[0005] In response to the technical problems existing in the prior art, the present invention provides a hepatobiliary CT image processing method and system, which solves the problems in the prior art of single hepatobiliary tissue evaluation, lack of nonlinear feature analysis, and neglect of the overall coordinated changes of the hepatobiliary system.

[0006] The technical solution of the present invention to solve the above technical problems is as follows: a hepatobiliary CT image processing method, comprising the following steps: acquiring a hepatobiliary CT image and establishing a three-dimensional model of hepatobiliary tissue, wherein the hepatobiliary CT image includes a plurality of two-dimensional slice images and a slice depth value of each two-dimensional slice image; performing feature recognition analysis on the three-dimensional hepatobiliary tissue model to obtain a hepatobiliary tissue feature data set; performing comprehensive analysis on the hepatobiliary tissue feature data set to obtain a liver tissue health index and a gallbladder tissue health index; performing comprehensive analysis on the liver tissue health index and the gallbladder tissue health index to obtain a comprehensive hepatobiliary tissue health index, performing judgment analysis on the liver tissue health index and the gallbladder tissue health index, and marking the hepatobiliary tissue as healthy based on the judgment analysis results; wherein the specific formula for calculating the comprehensive hepatobiliary tissue health index is as follows: ;in, is the comprehensive health index of liver and gallbladder tissue, is a natural constant, is the liver tissue health index, is the liver tissue weight coefficient stored in the database, is the bile tissue health index, is the bile tissue weight coefficient stored in the database, .

[0007] Furthermore, the two-dimensional slice image includes pixel values ​​and two-dimensional coordinates of several hepatobiliary pixel points, the hepatobiliary tissue feature data set includes a liver tissue feature data set and a gallbladder tissue feature data set, the liver tissue feature data set includes a geometric complexity index, a grayscale fluctuation index, and a metabolically active area index of the liver tissue area, and the gallbladder tissue feature data set includes a gallbladder wall thickness uniformity index, a gallbladder stone feature index, and a gallbladder grayscale change index of the gallbladder tissue area.

[0008] Furthermore, the specific steps of establishing a three-dimensional model of hepatobiliary tissue are as follows: each two-dimensional slice image is uniformly sized, and each two-dimensional slice image is arranged in descending order based on the slice depth value of each two-dimensional slice image; several two-dimensional slice images arranged in descending order are stacked, and slice image interpolation is performed based on a preset interpolation rule to obtain a three-dimensional model of hepatobiliary tissue, wherein the three-dimensional model of hepatobiliary tissue includes several hepatobiliary voxel points, and the voxel value and three-dimensional coordinate value of each hepatobiliary voxel point.

[0009] Furthermore, the specific steps of performing slice image interpolation processing based on the preset interpolation rules are as follows: based on the slice depth value of each two-dimensional slice image, the slice depth spacing between adjacent two-dimensional slice images is judged and analyzed with a preset slice depth spacing threshold; if the slice depth spacing between adjacent two-dimensional slice images is higher than the preset slice depth spacing threshold, an interpolated slice image is inserted between the adjacent two-dimensional slice images, and the interpolated slice image includes a number of interpolated liver and gallbladder pixel points.

[0010] Furthermore, the specific steps of performing feature recognition analysis on the three-dimensional model of hepatobiliary tissue to obtain a hepatobiliary tissue feature data set are as follows: grayscale processing is performed on the voxel value of each hepatobiliary voxel point in the three-dimensional model of hepatobiliary tissue to obtain the grayscale voxel value of each hepatobiliary voxel point in the three-dimensional model of hepatobiliary tissue; based on the preset liver tissue grayscale voxel interval and gallbladder tissue grayscale voxel interval, the grayscale voxel value of each hepatobiliary voxel point in the three-dimensional model of hepatobiliary tissue is binarized to obtain the liver tissue area and gallbladder tissue area in the three-dimensional model of hepatobiliary tissue, wherein the liver tissue area includes several liver voxel points and the grayscale voxel value of each liver voxel point, the bile tissue area includes several bile voxel points and the grayscale voxel value of each bile voxel point; the liver tissue areas are comprehensively analyzed to obtain the geometric complexity index, grayscale fluctuation index, and metabolically active area index of the liver tissue areas, and comprehensive analysis is performed to obtain the liver tissue health index; the bile tissue areas are comprehensively analyzed to obtain the gallbladder wall thickness uniformity index, gallbladder stone characteristic index, and gallbladder grayscale change index of the bile tissue areas, and comprehensive analysis is performed to obtain the bile tissue health index.

[0011] Furthermore, the specific formula for calculating the liver tissue health index is as follows: ;in, is the liver tissue health index, is the geometric complexity index of the liver tissue area, is the geometric coefficient of liver tissue stored in the database, is the grayscale fluctuation index of the liver tissue area, is the liver tissue grayscale fluctuation coefficient stored in the database, is the metabolically active area index of the liver tissue region, is the metabolic activity coefficient of liver tissue stored in the database.

[0012] Furthermore, the specific steps for obtaining the geometric complexity index, grayscale fluctuation index, and metabolically active area index of the liver tissue region are as follows: performing surface area analysis and boundary analysis on the liver tissue region respectively to obtain the liver tissue surface area value and liver tissue boundary length value of the liver tissue region, and performing comprehensive analysis to obtain the geometric complexity index of the liver tissue region; reading the grayscale voxel value of each liver voxel point in the liver tissue region, and performing comprehensive analysis to obtain the grayscale fluctuation index of the liver tissue region; comparing the grayscale voxel value of each liver voxel point in the liver tissue region with the preset metabolically active grayscale voxel value, and taking the liver voxel point with a grayscale voxel value higher than the preset metabolically active grayscale voxel value as the metabolically active liver voxel point, and performing comprehensive analysis to obtain the metabolically active area index of the liver tissue region.

[0013] Furthermore, the specific formula for calculating the bile tissue health index is as follows: ;in, is the bile tissue health index, is the gallbladder wall thickness uniformity index in the bile tissue area, is the bile tissue thickness uniformity coefficient stored in the database, is the gallstone characteristic index of the gallbladder tissue area, is the characteristic coefficient of gallstones stored in the database, is the gallbladder grayscale change index of the bile tissue area, is the grayscale variation coefficient of bile tissue stored in the database.

[0014] Furthermore, the specific steps for obtaining the gallbladder wall thickness uniformity index, gallbladder stone characteristic index, and gallbladder grayscale variation index of the gallbladder tissue region are as follows: for the gallbladder tissue region, arbitrarily select several gallbladder wall measurement points, and analyze the gallbladder wall thickness value of each gallbladder wall measurement point respectively; compare and analyze the gallbladder wall thickness value of each gallbladder wall measurement point to obtain the maximum gallbladder wall thickness and the minimum gallbladder wall thickness of the gallbladder tissue region; compare the gallbladder wall thickness value of each gallbladder wall measurement point, the maximum gallbladder wall thickness and the minimum gallbladder wall thickness of the gallbladder tissue region respectively. Comprehensive analysis was performed to obtain the uniformity index of gallbladder wall thickness in the bile tissue area. The grayscale voxel value of each bile voxel point in the bile tissue area was compared with the preset grayscale voxel value of gallbladder nodules. The bile voxel points with grayscale voxel values ​​higher than the preset grayscale voxel value of gallbladder nodules were taken as the characteristic voxel points of gallbladder stones, and the number of characteristic voxel points of gallbladder stones was taken as the characteristic index of gallbladder stones in the bile tissue area. The grayscale voxel value of each bile voxel point in the bile tissue area was read and comprehensively analyzed to obtain the grayscale fluctuation index of the bile tissue area.

[0015] A hepatobiliary CT image processing system comprises: a three-dimensional model establishment module, a feature recognition and analysis module, a health analysis module, a comprehensive analysis module, and a judgment and analysis module; the three-dimensional model establishment module is used to acquire a hepatobiliary CT image and establish a three-dimensional model of hepatobiliary tissue, wherein the hepatobiliary CT image comprises a plurality of two-dimensional slice images and a slice depth value of each two-dimensional slice image; the feature recognition and analysis module is used to perform feature recognition and analysis on the three-dimensional model of the hepatobiliary tissue to obtain a hepatobiliary tissue feature data set; the health analysis module is used to perform a comprehensive analysis on the hepatobiliary tissue feature data set to obtain a liver tissue health index and a gallbladder tissue health index; the comprehensive analysis module is used to perform a comprehensive analysis on the liver tissue health index and the gallbladder tissue health index to obtain a comprehensive hepatobiliary tissue health index; the judgment and analysis module is used to perform a judgment and analysis on the comprehensive hepatobiliary tissue health index and a preset comprehensive hepatobiliary tissue health assessment interval, and to mark the hepatobiliary tissue as healthy based on the judgment and analysis results.

[0016] The beneficial effects of the present invention are as follows: by constructing a complete three-dimensional model of hepatobiliary tissue and extracting multidimensional features, the method significantly improves the accuracy and comprehensiveness of hepatobiliary health assessment. First, by uniformly sizing, depth sorting, stacking, and interpolating two-dimensional slice images, the generated three-dimensional model retains the spatial structure and anatomical information of the hepatobiliary tissue, effectively eliminating the problem of data discontinuity between slices, and providing an accurate data basis for subsequent analysis. Second, based on the different physiological characteristics of hepatobiliary tissue, the geometric complexity index, grayscale fluctuation index, and metabolically active area index of liver tissue, as well as the wall thickness uniformity index, stone characteristic index, and grayscale variation index of the gallbladder are designed. These features comprehensively reflect the health status of hepatobiliary tissue from multiple dimensions of morphology, density, and function, and can sensitively capture early signals of complex lesions. Finally, by nonlinearly integrating the liver tissue health index and the gallbladder tissue health index, the calculated hepatobiliary tissue comprehensive health index can comprehensively assess the overall health status of the hepatobiliary system and flexibly adapt to clinical needs under different pathological conditions. This method has demonstrated significant practicality and scientific value in early disease screening, diagnosis, and treatment plan formulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of a method for processing hepatobiliary CT images according to the present invention.

[0018] Figure 2 This is a flowchart of the specific steps for establishing a three-dimensional model of liver and gallbladder tissue in a liver and gallbladder CT image processing method of the present invention.

[0019] Figure 3 This is a block diagram of a hepatobiliary CT image processing system of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0021] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the described features. In the description of this application, "plurality" means two or more, unless otherwise specifically specified.

[0022] In the description of this application, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art will recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in this application.

[0023] The overall approach to the problems in the embodiments of this application is as follows:

[0024] First, hepatobiliary CT images were acquired and a three-dimensional model of hepatobiliary tissue was established. This involved uniformly resizing the two-dimensional slice images, sorting and stacking them by slice depth values, and generating a three-dimensional model based on preset interpolation rules. The model included hepatobiliary voxel points, their voxel values, and three-dimensional coordinate values.

[0025] Next, feature recognition analysis is performed on the three-dimensional model to separate the liver tissue and bile tissue regions: this includes dividing the three-dimensional model into liver tissue regions and bile tissue regions through grayscale value processing and binarization, and extracting voxel data for each region, including grayscale voxel value, voxel position, etc.

[0026] Finally, comprehensive feature analysis was performed on the liver tissue and bile tissue regions respectively: including extracting the geometric complexity index, grayscale fluctuation index, and metabolically active area index from the liver tissue region, and calculating the liver tissue health index; extracting the gallbladder wall thickness uniformity index, gallbladder stone characteristic index, and grayscale change index from the bile tissue region, and calculating the bile tissue health index. By comprehensively analyzing the liver tissue health index and bile tissue health index, the comprehensive health index of the liver and bile tissues was calculated and evaluated.

[0027] See also Figure 1 The embodiment of the present invention provides a technical solution: a method for processing hepatobiliary CT images, comprising the following steps: acquiring a hepatobiliary CT image and establishing a three-dimensional model of hepatobiliary tissue, wherein the hepatobiliary CT image includes a plurality of two-dimensional slice images and a slice depth value of each two-dimensional slice image (i.e., a Z coordinate value, with the slice direction as the Z coordinate axis and the first slice position as the coordinate axis origin); performing feature recognition analysis on the three-dimensional model of the hepatobiliary tissue to obtain a hepatobiliary tissue feature data set; performing comprehensive analysis on the hepatobiliary tissue feature data set to obtain a liver tissue health index and a gallbladder tissue health index; and comprehensively analyzing the liver tissue health index and the gallbladder tissue health index to obtain a hepatobiliary tissue health index. The liver and gallbladder tissue comprehensive health index is compared with the preset liver and gallbladder tissue comprehensive health assessment range, and the liver and gallbladder tissue is marked as healthy according to the judgment and analysis results. The specific process is as follows: if the liver and gallbladder tissue comprehensive health index is within the preset liver and gallbladder tissue comprehensive health assessment range, the liver tissue area and the gallbladder tissue area are marked as normal; if the liver and gallbladder tissue comprehensive health index is outside the preset liver and gallbladder tissue comprehensive health assessment range, it is judged whether the liver tissue health index and the gallbladder tissue health index are respectively within the preset liver tissue health assessment range and the gallbladder tissue health assessment range, and the health index outside the assessment range is regarded as an abnormal area and marked.

[0028] The specific formula for calculating the comprehensive health index of hepatobiliary tissue is as follows: ;in, is the comprehensive health index of liver and gallbladder tissue, is a natural constant, and in this embodiment, its value is 2.718. is the liver tissue health index, is the liver tissue weight coefficient stored in the database, is the bile tissue health index, is the bile tissue weight coefficient stored in the database, .

[0029] The implementation example of calculating the comprehensive health index of liver and gallbladder tissue is as follows, and the existing data is as follows:

[0030] The liver tissue health index is: 0.824.

[0031] The liver tissue weight coefficient stored in the database is: 0.68.

[0032] The bile tissue health index is: 0.645.

[0033] The bile tissue weight coefficient stored in the database is 0.32.

[0034] The natural constant is: 2.718.

[0035] Substituting the above data into the specific formula for calculating the comprehensive health index of hepatobiliary tissue, we obtain:

[0036] Comprehensive health index of liver and gallbladder tissue = 1 / (1+e -（0.68*0.824+0.32*0.645） )≈0683.

[0037] The two-dimensional slice image includes the pixel values ​​and two-dimensional coordinates of several hepatobiliary pixel points. The hepatobiliary tissue feature dataset includes the liver tissue feature dataset and the gallbladder tissue feature dataset. The liver tissue feature dataset includes the geometric complexity index, grayscale fluctuation index, and metabolically active area index of the liver tissue region. The gallbladder tissue feature dataset includes the gallbladder wall thickness uniformity index, gallstone feature index, and gallbladder grayscale variation index of the gallbladder tissue region.

[0038] Specifically, such as Figure 2 As shown, the specific steps of establishing a three-dimensional model of hepatobiliary tissue are as follows: each two-dimensional slice image is uniformly sized, and each two-dimensional slice image is arranged in descending order based on the slice depth value of each two-dimensional slice image; several two-dimensional slice images arranged in descending order are stacked, and slice image interpolation is performed based on a preset interpolation rule to obtain a three-dimensional model of hepatobiliary tissue, which includes several hepatobiliary voxel points, and the voxel value and three-dimensional coordinate value of each hepatobiliary voxel point.

[0039] The specific steps of performing slice image interpolation processing based on a preset interpolation rule are as follows: based on the slice depth value of each two-dimensional slice image, the slice depth spacing between adjacent two-dimensional slice images is judged and analyzed with a preset slice depth spacing threshold; if the slice depth spacing between adjacent two-dimensional slice images is higher than the preset slice depth spacing threshold, an interpolated slice image is inserted between the adjacent two-dimensional slice images, and the interpolated slice image includes a number of interpolated liver and gallbladder pixel points.

[0040] The specific formula for calculating the pixel value of each interpolated hepatobiliary pixel in the interpolated slice image is as follows: ;in, is the interpolated slice image Interpolated pixel values ​​of liver and gallbladder pixels, is the first interpolation coefficient stored in the database, is the second interpolation coefficient stored in the database, , is the first two-dimensional slice image in the adjacent two-dimensional slice image. The pixel value of the liver and gallbladder pixels, is the first two-dimensional slice image in the second adjacent two-dimensional slice image. The pixel value of the liver and gallbladder pixels, =1, 2, 3, ..., , is the number of interpolated liver and gallbladder pixels in the interpolated slice image.

[0041] In this implementation, by arranging the depth values ​​of the two-dimensional slice images in descending order and stacking them, the tomographic data of the hepatobiliary tissue are orderly combined, and reconstruction from two-dimensional to three-dimensional is achieved, the spatial structure and anatomical information of the hepatobiliary tissue is retained, and a complete three-dimensional data foundation is provided for subsequent feature analysis. The introduction of interpolation rules solves the problem of excessive depth spacing between adjacent slices. By inserting interpolated slice images, the gaps between data are filled, and the three-dimensional model has higher continuity and smoothness between slices, thereby improving the fineness of the three-dimensional model. By unifying the size of each two-dimensional slice image, slice data of different resolutions or pixel sizes can be analyzed for consistency, enhancing the universality and flexibility of the algorithm. , which can adapt to different CT devices and image formats. The calculation of each pixel value in the interpolated slice is based on the pixel value of the adjacent slice and the preset interpolation coefficient, and is calculated by the formula, so that the interpolated image can more accurately reflect the actual tissue density changes, laying a data foundation for the accurate extraction and analysis of liver and gallbladder regional features. The complete and continuous three-dimensional model provides reliable data support for subsequent feature recognition (such as liver metabolic areas, gallbladder stone characteristics, etc.) and health assessment, and can more accurately reflect the overall health status of liver and gallbladder tissue. By dynamically adjusting the interpolation rules (such as depth spacing threshold and interpolation coefficient), the three-dimensional model can be flexibly optimized according to different clinical needs, thereby better supporting personalized diagnosis and treatment planning.

[0042] Specifically, the specific steps of performing feature recognition analysis on the three-dimensional model of hepatobiliary tissue to obtain the hepatobiliary tissue feature data set are as follows: grayscale processing is performed on the voxel value of each hepatobiliary voxel point in the three-dimensional model of hepatobiliary tissue to obtain the grayscale voxel value of each hepatobiliary voxel point in the three-dimensional model of hepatobiliary tissue; based on the preset liver tissue grayscale voxel interval and gallbladder tissue grayscale voxel interval, the grayscale voxel value of each hepatobiliary voxel point in the three-dimensional model of hepatobiliary tissue is binarized to obtain the liver tissue area and gallbladder tissue area in the three-dimensional model of hepatobiliary tissue. The liver tissue area includes several The liver voxel points and the grayscale voxel value of each liver voxel point are included in the bile tissue area, and the grayscale voxel value of each bile voxel point is included in the bile tissue area. A comprehensive analysis is performed on the liver tissue area to obtain the geometric complexity index, grayscale fluctuation index, and metabolically active area index of the liver tissue area, and a comprehensive analysis is performed to obtain the liver tissue health index. A comprehensive analysis is performed on the bile tissue area to obtain the gallbladder wall thickness uniformity index, gallbladder stone characteristic index, and gallbladder grayscale change index of the bile tissue area, and a comprehensive analysis is performed to obtain the bile tissue health index.

[0043] The specific formula for calculating the liver tissue health index is as follows: ;in, is the liver tissue health index, is the geometric complexity index of the liver tissue area, is the geometric coefficient of liver tissue stored in the database, is the grayscale fluctuation index of the liver tissue area, is the liver tissue grayscale fluctuation coefficient stored in the database, is the metabolically active area index of the liver tissue region, is the metabolic activity coefficient of liver tissue stored in the database.

[0044] The specific formula for calculating the bile tissue health index is as follows: ;in, is the bile tissue health index, is the gallbladder wall thickness uniformity index in the bile tissue area, is the bile tissue thickness uniformity coefficient stored in the database, is the gallstone characteristic index of the gallbladder tissue area, is the characteristic coefficient of gallstones stored in the database, is the gallbladder grayscale change index of the bile tissue area, is the grayscale variation coefficient of bile tissue stored in the database.

[0045] In this embodiment, by grayscale processing and regional segmentation of voxel points in the three-dimensional model of hepatobiliary tissue, the core features of the liver tissue region and the gallbladder tissue region are extracted respectively, which comprehensively reflects the health status of the hepatobiliary tissue. The feature extraction process is meticulous and multidimensional, and can accurately quantify the geometric morphology, density distribution and functional characteristics of the liver and gallbladder, which helps to comprehensively evaluate the health level of the hepatobiliary tissue. The geometric complexity index, grayscale fluctuation index and metabolically active area index of the liver tissue focus on reflecting the structural integrity, density uniformity and distribution characteristics of the functional activity area of ​​the liver. The gallbladder wall thickness uniformity index, gallstone characteristic index and grayscale change index of the gallbladder tissue focus on the structural stability, pathological characteristics and internal density changes of the gallbladder. This targeted feature analysis design ensures that the health index of each tissue can accurately reflect its characteristics and possible lesions. By introducing multiple features and taking the comprehensive analysis of features as the core, the health index of liver tissue and gallbladder tissue is expanded from a single feature evaluation to a multi-dimensional comprehensive evaluation. The calculation formula of the health index not only takes into account the weight of each feature, but also enhances the accuracy of the evaluation results through the relationship between features, avoiding the misjudgment that may be caused by a single feature. The calculation formula of the health index combines each feature with its importance coefficient, making the model scientific and flexible: through the weight coefficients such as the geometric morphology coefficient and the grayscale fluctuation coefficient stored in the database, it ensures that the contribution of each feature to the health index is in line with actual medical significance; in different application scenarios, the weight coefficient can be adjusted according to specific needs to adapt to different This method can meet different evaluation needs and improve the applicability of the evaluation model. By refining the feature analysis of liver tissue and bile tissue, this method can capture early lesions in the tissue (such as liver fibrosis, fatty liver, cholecystitis or stone formation) and their potential health risks. By quantifying the health index, it provides a scientific basis for early diagnosis, which helps doctors to accurately formulate treatment plans and optimize treatment effects. This health index calculation method based on multidimensional features can efficiently process large-scale hepatobiliary CT image data, facilitate batch analysis and health screening in clinical practice, and significantly improve the efficiency and practicality of medical image analysis. Through the construction of a three-dimensional model of hepatobiliary tissue, automatic feature extraction and health index calculation, this method provides a technical basis for the automated analysis and intelligent diagnosis of medical images, which helps to promote the development of medical technology.

[0046] Specifically, the specific steps for obtaining the geometric complexity index, grayscale fluctuation index, and metabolically active area index of the liver tissue region are as follows: surface area analysis and boundary analysis are performed on the liver tissue region respectively (the surface mesh of the liver can be obtained by extracting boundary points, i.e., the contact points between the liver and the surrounding tissues, from the three-dimensional voxel model of the liver. Then, the area calculation formula of the triangular mesh is used, i.e., the area is equal to half of the cross product value of the two edge vectors of the triangle, and the sum of the areas of these surface triangles is calculated to obtain the surface area of ​​the liver. At the same time, the boundary length is obtained by extracting the edge contour of the liver region on each two-dimensional slice and calculating the distance between these contour points. The boundary lengths of all slices are accumulated to obtain the total boundary length of the liver), the liver tissue surface area value and the liver tissue boundary length value of the liver tissue region are obtained, and a comprehensive analysis is performed to obtain the geometric complexity index of the liver tissue region; the grayscale voxel value of each liver voxel point in the liver tissue region is read and a comprehensive analysis is performed to obtain the grayscale fluctuation index of the liver tissue region; the grayscale voxel value of each liver voxel point in the liver tissue region is compared and analyzed with the preset metabolically active grayscale voxel value, and the liver voxel points with grayscale voxel values ​​higher than the preset metabolically active grayscale voxel value are regarded as metabolically active liver voxel points, and a comprehensive analysis is performed to obtain the metabolically active regional index of the liver tissue region.

[0047] The specific formulas for calculating the geometric complexity index, grayscale fluctuation index, and metabolically active area index of the liver tissue region are as follows: ;in, is the geometric complexity index of the liver tissue area, is the liver tissue surface area value of the liver tissue region, is the liver tissue boundary length value of the liver tissue area, is the grayscale fluctuation index of the liver tissue area, The first The grayscale voxel value of the liver voxel point, is the metabolically active area index of the liver tissue region, The first Grayscale voxel value of metabolically active liver voxel point, =1, 2, 3, ..., , is the number of liver voxels in the liver tissue area, =1, 2, 3, ..., , It is the number of metabolically active liver voxels in the liver tissue area.

[0048] In this embodiment, the geometric morphological characteristics of liver tissue are accurately characterized through surface area analysis and boundary length analysis. The surface area value of liver tissue reflects the complexity of its external contour, and the boundary length value reveals the morphological changes of liver tissue on two-dimensional slices. These geometric characteristics are further integrated into a geometric morphological complexity index, which can effectively identify whether the liver has morphological abnormalities (such as surface irregularities caused by liver fibrosis or cirrhosis). By reading the grayscale value of liver voxel points and calculating the grayscale fluctuation index, the fluctuation of regional density distribution of liver tissue is quantified. Healthy liver tissue usually has uniform density, while lesions such as fatty liver and inflammation can cause a significant increase in density fluctuation. This feature can sensitively reflect the density changes inside liver tissue, which helps to detect early lesions, especially some hidden lesions that are difficult to identify with the naked eye. By comparing the voxel grayscale value with the preset metabolically active grayscale threshold, the metabolically active areas of liver tissue are identified, and the metabolically active area index is calculated. The size and distribution of can reflect the functional status of the liver, help identify local metabolic abnormalities (such as reduced local activity caused by inflammation, tumors or insufficient blood flow), and provide a more functional reference for disease diagnosis. The geometric complexity index is calculated through a comprehensive analysis of surface area and boundary length. It can balance the complexity of the external morphology and internal contour of liver tissue and adapt to various pathological morphologies. The grayscale fluctuation index adopts a combination of point-by-point analysis and comprehensive evaluation to sensitively capture the distribution characteristics of grayscale values, avoiding local abnormalities that may be missed by simple averaging. The metabolically active area index is based on the screening and comprehensive analysis of active voxel points, directly quantifying the activity of metabolic function, and is particularly suitable for reflecting the dynamic functional status of the liver. Through the extraction and quantitative analysis of multidimensional features, this part of the method can reflect the health status of liver tissue from multiple dimensions of morphology, density and function. Compared with traditional single indicator evaluation, this method can provide richer information and improve the accuracy and comprehensiveness of health assessment.

[0049] Specifically, the specific steps for obtaining the gallbladder wall thickness uniformity index, gallbladder stone characteristic index, and gallbladder grayscale variation index of the gallbladder tissue region are as follows: for the gallbladder tissue region, arbitrarily select several gallbladder wall measurement points, and analyze the gallbladder wall thickness value of each gallbladder wall measurement point (i.e., the distance value between the outer surface point and the inner surface point); compare and analyze the gallbladder wall thickness value of each gallbladder wall measurement point to obtain the maximum gallbladder wall thickness and the minimum gallbladder wall thickness of the gallbladder tissue region; compare the gallbladder wall thickness value of each gallbladder wall measurement point, the maximum gallbladder wall thickness and the minimum gallbladder wall thickness of the gallbladder tissue region The minimum value of the wall thickness was comprehensively analyzed to obtain the uniformity index of the gallbladder wall thickness in the bile tissue area; the grayscale voxel value of each bile voxel point in the bile tissue area was compared and analyzed with the preset grayscale voxel value of the gallbladder nodule, and the bile voxel point with a grayscale voxel value higher than the preset grayscale voxel value of the gallbladder nodule was regarded as the characteristic voxel point of the gallbladder stone, and the number of the characteristic voxel points of the gallbladder stone was regarded as the characteristic index of the gallbladder stone in the bile tissue area; the grayscale voxel value of each bile voxel point in the bile tissue area was read and comprehensively analyzed to obtain the grayscale fluctuation index of the bile tissue area.

[0050] The specific formulas for calculating the gallbladder wall thickness uniformity index and gallbladder grayscale variation index in the gallbladder tissue area are as follows: ;in, is the gallbladder wall thickness uniformity index in the bile tissue area, is the maximum gallbladder wall thickness in the bile tissue area, is the minimum gallbladder wall thickness in the bile tissue area, The first The gallbladder wall thickness value at each gallbladder wall measurement point, is the gallbladder grayscale change index of the bile tissue area, The first The grayscale voxel value of each bile voxel point, =1, 2, 3, ..., , is the number of measurement points on the gallbladder wall, =1, 2, 3, ..., , is the number of bile duct points in the bile tissue area.

[0051] In this embodiment, by selecting several gallbladder wall measurement points and analyzing the wall thickness value of each point, the thickness distribution characteristics of the gallbladder wall can be fully reflected. The uniformity of the gallbladder wall thickness is closely related to the health status, and unevenness may indicate cholecystitis or other lesions. The maximum and minimum values ​​of the gallbladder wall thickness are calculated, and the gallbladder wall thickness uniformity index is obtained through comprehensive analysis. It can quantify the variation range of the gallbladder wall thickness distribution and help identify the lesion area. This feature can sensitively reflect the early pathological changes of the gallbladder wall, such as local thickening caused by cholecystitis, and provide a scientific basis for the early diagnosis of the disease. By comparing the grayscale value of each voxel in the gallbladder area with the preset stone grayscale threshold, high-density stone areas are identified, and the number of stone characteristic voxel points is used as the gallbladder stone characteristic index. The gallbladder stone characteristic index can directly quantify the severity of the stone (such as the number of stones) and provide a quantitative basis for further treatment (such as whether surgery is required). This method is based on simple and intuitive voxel grayscale comparison and is efficient and accurate in calculation. The grayscale variation index is calculated by reading the grayscale value of each voxel in the gallbladder region and performing a comprehensive analysis. A healthy gallbladder typically has a relatively uniform grayscale distribution, whereas cholecystitis, abnormal bile concentration, or other pathologies may result in abnormal grayscale distribution. The grayscale variation index quantifies the degree of grayscale fluctuation, reflecting the health of the bile in the gallbladder and helping to identify density unevenness caused by abnormal bile flow or inflammation. This feature can capture subtle abnormalities in the gallbladder from the perspective of density changes. Combined with other features, it can comprehensively reflect the health of the gallbladder. By extracting features from three dimensions: gallbladder wall thickness, stone characteristics, and grayscale variation, it covers the three core health indicators of gallbladder morphology, pathology, and function. Different features complement each other, allowing a comprehensive assessment of gallbladder health from multiple perspectives, avoiding misjudgments that may result from a single indicator. This multidimensional analysis method greatly improves the accuracy and comprehensiveness of gallbladder health assessment and is suitable for the precise diagnosis of complex lesions.

[0052] See also Figure 3A hepatobiliary CT image processing system includes: a three-dimensional model building module, a feature recognition and analysis module, a health analysis module, a comprehensive analysis module, and a judgment analysis module; the three-dimensional model building module is used to obtain hepatobiliary CT images and establish a three-dimensional model of hepatobiliary tissue, where the hepatobiliary CT images include several two-dimensional slice images and a slice depth value of each two-dimensional slice image; the feature recognition and analysis module is used to perform feature recognition and analysis on the three-dimensional model of hepatobiliary tissue to obtain a hepatobiliary tissue feature data set; the health analysis module is used to perform comprehensive analysis on the hepatobiliary tissue feature data set to obtain a liver tissue health index and a gallbladder tissue health index; the comprehensive analysis module is used to perform comprehensive analysis on the liver tissue health index and the gallbladder tissue health index to obtain a comprehensive hepatobiliary tissue health index; the judgment analysis module is used to perform judgment analysis on the comprehensive hepatobiliary tissue health index and a preset comprehensive hepatobiliary tissue health assessment interval, and mark the hepatobiliary tissue as healthy according to the judgment analysis results.

[0053] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0054] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0055] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0056] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0057] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0058] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0059] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for processing hepatobiliary CT images, characterized in that: The following steps are involved: Acquire a hepatobiliary CT image and establish a three-dimensional model of hepatobiliary tissue, wherein the hepatobiliary CT image includes a plurality of two-dimensional slice images and a slice depth value of each two-dimensional slice image, and the two-dimensional slice image includes pixel values ​​and two-dimensional coordinates of a plurality of hepatobiliary pixel points; Performing feature recognition analysis on the three-dimensional model of hepatobiliary tissue to obtain a hepatobiliary tissue feature dataset, including a liver tissue feature dataset and a gallbladder tissue feature dataset. The liver tissue feature dataset includes a geometric complexity index, a grayscale fluctuation index, and a metabolically active area index of the liver tissue region; the gallbladder tissue feature dataset includes a gallbladder wall thickness uniformity index, a gallbladder stone feature index, and a gallbladder grayscale variation index of the gallbladder tissue region; Comprehensive analysis of the liver and gallbladder tissue feature dataset was performed to obtain the liver tissue health index and gallbladder tissue health index; Comprehensively analyze the liver tissue health index and gallbladder tissue health index to obtain the liver and gallbladder tissue comprehensive health index, and perform judgment analysis with the preset liver and gallbladder tissue comprehensive health assessment interval, and mark the liver and gallbladder tissue health based on the judgment and analysis results; The specific formula for calculating the comprehensive health index of hepatobiliary tissue is as follows: Wherein, GdJ is the comprehensive health index of hepatobiliary tissue, e is a natural constant, GzZ is the liver tissue health index, μ1 is the liver tissue weight coefficient stored in the database, DzZ is the gallbladder tissue health index, μ2 is the gallbladder tissue weight coefficient stored in the database, μ1+μ2=1; The specific steps for performing feature recognition analysis on the three-dimensional model of liver and gallbladder tissue to obtain the liver and gallbladder tissue feature dataset are as follows: grayscale processing is performed on the voxel value of each hepatobiliary voxel point in the three-dimensional model of hepatobiliary tissue to obtain the grayscale voxel value of each hepatobiliary voxel point in the three-dimensional model of hepatobiliary tissue; Based on the preset liver tissue grayscale voxel interval and the bile tissue grayscale voxel interval, the grayscale voxel value of each hepatobiliary voxel point in the hepatobiliary tissue three-dimensional model is binarized to obtain a liver tissue region and a bile tissue region in the hepatobiliary tissue three-dimensional model, wherein the liver tissue region includes a plurality of liver voxel points and the grayscale voxel value of each liver voxel point, and the bile tissue region includes a plurality of bile voxel points and the grayscale voxel value of each bile voxel point; Comprehensive analysis was performed on the liver tissue regions to obtain the geometric complexity index, grayscale fluctuation index, and metabolically active area index of the liver tissue regions, and comprehensive analysis was performed to obtain the liver tissue health index; Comprehensive analysis was performed on the bile tissue area to obtain the gallbladder wall thickness uniformity index, gallbladder stone characteristic index, and gallbladder grayscale change index of the bile tissue area, and comprehensive analysis was performed to obtain the bile tissue health index; The specific steps for obtaining the geometric complexity index, grayscale fluctuation index, and metabolically active area index of the liver tissue region are as follows: Surface area analysis and boundary analysis were performed on the liver tissue region to obtain the liver tissue surface area value and liver tissue boundary length value of the liver tissue region, and comprehensive analysis was performed to obtain the geometric complexity index of the liver tissue region; Read the grayscale voxel value of each liver voxel point in the liver tissue area and perform comprehensive analysis to obtain the grayscale fluctuation index of the liver tissue area; The grayscale voxel value of each liver voxel point in the liver tissue area is compared and analyzed with the preset metabolically active grayscale voxel value, and the liver voxel point with a grayscale voxel value higher than the preset metabolically active grayscale voxel value is regarded as a metabolically active liver voxel point. A comprehensive analysis is then performed to obtain the metabolically active regional index of the liver tissue area; The specific steps for obtaining the gallbladder wall thickness uniformity index, gallstone characteristic index, and gallbladder grayscale variation index of the gallbladder tissue region are as follows: For the bile tissue area, several gallbladder wall measurement points were randomly selected, and the gallbladder wall thickness value of each gallbladder wall measurement point was analyzed separately; Compare and analyze the gallbladder wall thickness values ​​at each gallbladder wall measurement point to obtain the maximum and minimum gallbladder wall thickness values ​​in the gallbladder tissue area; The gallbladder wall thickness uniformity index of the gallbladder tissue area was obtained by comprehensively analyzing the gallbladder wall thickness value of each gallbladder wall measurement point, the maximum gallbladder wall thickness value of the gallbladder tissue area, and the minimum gallbladder wall thickness value of the gallbladder tissue area. The grayscale voxel value of each bile voxel point in the bile tissue area is compared with the preset grayscale voxel value of gallstones. The bile voxel points with grayscale voxel values ​​higher than the preset grayscale voxel value of gallstones are regarded as characteristic voxel points of gallstones, and the number of characteristic voxel points of gallstones is regarded as the characteristic index of gallstones in the bile tissue area. The grayscale voxel value of each bile voxel point in the bile tissue area is read and comprehensively analyzed to obtain the grayscale fluctuation index of the bile tissue area.

2. The hepatobiliary CT image processing method according to claim 1, characterized in that: The specific steps for establishing a three-dimensional model of hepatobiliary tissue are as follows: Performing a uniform size processing on each two-dimensional slice image, and performing a descending sorting processing on each two-dimensional slice image based on a slice depth value of each two-dimensional slice image; The plurality of two-dimensional slice images arranged in descending order are stacked, and the slice images are interpolated based on a preset interpolation rule to obtain a three-dimensional model of the hepatobiliary tissue. The three-dimensional model of the hepatobiliary tissue includes a plurality of hepatobiliary voxel points, and a voxel value and a three-dimensional coordinate value of each hepatobiliary voxel point.

3. The hepatobiliary CT image processing method according to claim 2, characterized in that: The specific steps of performing slice image interpolation processing based on the preset interpolation rules are as follows: Based on the slice depth value of each two-dimensional slice image, the slice depth interval between adjacent two-dimensional slice images is judged and analyzed with a preset slice depth interval threshold; If the slice depth spacing between adjacent two-dimensional slice images is higher than a preset slice depth spacing threshold, an interpolated slice image is inserted between the adjacent two-dimensional slice images, where the interpolated slice image includes a plurality of interpolated hepatobiliary pixel points.

4. The hepatobiliary CT image processing method according to claim 1, characterized in that: The specific formula for calculating the liver tissue health index is as follows: Among them, GzZ is the liver tissue health index, JhX is the geometric complexity index of the liver tissue area, ξ1 is the liver tissue geometric coefficient stored in the database, HdB is the grayscale fluctuation index of the liver tissue area, ξ2 is the liver tissue grayscale fluctuation coefficient stored in the database, DxH is the metabolically active area index of the liver tissue area, and ξ3 is the liver tissue metabolic activity coefficient stored in the database.

5. The hepatobiliary CT image processing method according to claim 1, characterized in that: The specific formula for calculating the bile tissue health index is as follows: Among them, DzZ is the gallbladder tissue health index, DbJ is the gallbladder wall thickness uniformity index of the gallbladder tissue area, η1 is the gallbladder tissue thickness uniformity coefficient stored in the database, DjT is the gallbladder stone characteristic index of the gallbladder tissue area, η2 is the gallbladder stone characteristic coefficient stored in the database, DhB is the gallbladder grayscale change index of the gallbladder tissue area, and η3 is the gallbladder tissue grayscale change coefficient stored in the database.

6. A hepatobiliary CT image processing system, applying the hepatobiliary CT image processing method according to any one of claims 1 to 5, characterized in that: include: 3D model building module, feature recognition and analysis module, health analysis module, comprehensive analysis module, and judgment analysis module; The three-dimensional model building module is used to obtain a hepatobiliary CT image and build a three-dimensional model of the hepatobiliary tissue, wherein the hepatobiliary CT image includes a plurality of two-dimensional slice images and a slice depth value of each two-dimensional slice image; The feature recognition and analysis module is used to perform feature recognition and analysis on the three-dimensional model of hepatobiliary tissue to obtain a hepatobiliary tissue feature data set; The health analysis module is used to perform a comprehensive analysis on the liver and gallbladder tissue feature data set to obtain a liver tissue health index and a gallbladder tissue health index; The comprehensive analysis module is used to comprehensively analyze the liver tissue health index and the gallbladder tissue health index to obtain the liver and gallbladder tissue comprehensive health index; The judgment and analysis module is used to judge and analyze the comprehensive health index of the hepatobiliary tissue and the preset comprehensive health assessment interval of the hepatobiliary tissue, and to mark the health of the hepatobiliary tissue according to the judgment and analysis results.

Images