Method and apparatus for calculating tissue structure measurement parameters, and electronic device

Abstract

This application provides a method, apparatus, and electronic device for calculating tissue structure measurement parameters, relating to the field of image processing technology. The method includes: acquiring computed tomography (CTA) angiography image data of a target region; wherein the CTA image data includes multiple two-dimensional slice images; segmenting the target tissue structure in the multiple two-dimensional slice images to obtain two-dimensional segmentation boundaries of the target tissue structure in each two-dimensional slice image; performing three-dimensional reconstruction based on the two-dimensional segmentation boundaries of the target tissue structure in each two-dimensional slice image to generate a three-dimensional geometric model of the target tissue structure; extracting geometric features of the target tissue structure from the three-dimensional geometric model, and calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features. The method provided by this application overcomes the deficiency of traditional two-dimensional measurements in failing to accurately reflect the spatial morphology of complex tissue structures, thereby significantly improving the accuracy and consistency of measurements.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, apparatus, and electronic device for calculating tissue structure measurement parameters. Background Technology

[0002] In the field of medical image analysis, precise geometric parameter measurement of complex soft tissue structures such as the mitral valve, aorta, blood vessels, and chordae tendineae is a key basis for the diagnosis of cardiovascular diseases, preoperative planning, and efficacy evaluation.

[0003] Currently, in clinical practice, most of these measurements still rely on ultrasound imaging or manual operation based on two-dimensional slices from computed tomography angiography (CTA). This involves physicians manually marking anatomical landmarks on the two-dimensional image plane and then estimating parameters such as the length, thickness, area, and major and minor axes of the target structure through experience or simple linear measurements. However, ultrasound imaging is often affected by factors such as acoustic window limitations, probe angle deviations, and tissue obstruction, resulting in unclear boundaries of subvalvular structures, calcified areas, and complex tissue junctions, leading to significant measurement errors. Furthermore, two-dimensional image-based measurements cannot accurately reproduce the curved, twisted, or irregularly oriented morphologies of tissues; linear distance measurements fail to reflect the actual length and three-dimensional contours of structures, and there is a lack of a unified, automated multi-dimensional parameter extraction mechanism for strip-like, tubular, or sheet-like structures. In addition, the measurement process, heavily reliant on human experience, is not only inefficient but also lacks consistency among different operators, severely hindering the development of precision medicine and intelligent assisted diagnosis and treatment.

[0004] Therefore, a solution is urgently needed to address the above problems. Summary of the Invention

[0005] This application provides a method, apparatus, and electronic device for calculating tissue structure measurement parameters to address the deficiencies in the prior art.

[0006] This application provides a method for calculating tissue structure measurement parameters, the method comprising: Acquire computed tomography (CTA) image data of the target area; wherein the CTA image data includes multiple two-dimensional slice images; The target tissue structure in the plurality of two-dimensional slice images is segmented to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image; Based on the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image, a three-dimensional reconstruction is performed to generate a three-dimensional geometric model of the target tissue structure. Geometric features of the target tissue structure are extracted from the three-dimensional geometric model, and at least one type of geometric measurement parameters of the target tissue structure are calculated based on the geometric features.

[0007] According to an embodiment of this application, a method for calculating tissue structure measurement parameters includes segmenting the target tissue structure in the plurality of two-dimensional slice images to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image, comprising: Each of the plurality of two-dimensional slice images is preprocessed to obtain a preprocessed image; A preset strategy is applied to the preprocessed image to extract the contour, and the closed contour of the target tissue structure is used as the two-dimensional segmentation boundary. The preset strategies include threshold segmentation, region growing, edge detection, or deep learning segmentation networks.

[0008] According to an embodiment of this application, a method for calculating tissue structure measurement parameters includes, in which three-dimensional reconstruction is performed based on the two-dimensional segmentation boundaries of the target tissue structure in each two-dimensional slice image to generate a three-dimensional geometric model of the target tissue structure, comprising: Establish the spatial coordinate relationship between the multiple two-dimensional slice images; Based on the spatial coordinate relationship, the two-dimensional segmentation boundaries of adjacent slices are interpolated and connected, and a three-dimensional geometric model composed of point clouds or meshes is generated by the surface reconstruction algorithm.

[0009] A method for calculating tissue structure measurement parameters is provided according to an embodiment of this application. The geometric features include at least one of the center path, normal direction, and boundary profile; The geometric measurement parameters include at least one of length, thickness, major diameter, minor diameter, endpoint distance, two-dimensional area, and three-dimensional surface area.

[0010] According to an embodiment of this application, a method for calculating tissue structure measurement parameters includes extracting geometric features of the target tissue structure from the three-dimensional geometric model and calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features, comprising: The center path is extracted from the three-dimensional geometric model using a skeleton extraction algorithm; By performing discrete curve integration on the central path, the spatial distances between adjacent points on the central path are calculated and accumulated to obtain the length of the target tissue structure.

[0011] According to an embodiment of this application, a method for calculating tissue structure measurement parameters is provided, wherein calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: Multiple test points are selected on the surface of the three-dimensional geometric model, and the corresponding normal direction is determined for each of the multiple test points. Along the normal direction at the point to be measured, find the intersection point of the normal direction and the other side surface of the three-dimensional geometric model, and take it as the corresponding point; Calculate the spatial distance between the point to be measured and the corresponding point, and use it as the local thickness at the point to be measured; The thickness distribution of the target tissue structure is obtained based on the local thickness of the multiple test points.

[0012] According to an embodiment of this application, a method for calculating tissue structure measurement parameters is provided, wherein calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: Ellipse fitting or principal direction analysis is performed on the boundary contour to determine the major and minor axes of the target tissue structure. The length of the major axis and the length of the minor axis are respectively used as the major axis and minor axis of the target tissue structure.

[0013] According to an embodiment of this application, a method for calculating tissue structure measurement parameters is provided, wherein calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: The endpoint positions of the target tissue structure are identified based on the curvature changes or endpoint features of the boundary contour. Calculate the spatial distance between the identified endpoints as the endpoint distance of the target organizational structure.

[0014] According to an embodiment of this application, a method for calculating tissue structure measurement parameters is provided, wherein calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: The two-dimensional area of ​​the target tissue structure is obtained by integrating the closed contour formed by the two-dimensional segmentation boundary. Based on the mesh representation of the three-dimensional geometric model, the areas of the triangular facets constituting the surface are accumulated to obtain the three-dimensional surface area of ​​the target tissue structure.

[0015] This application embodiment also provides a device for calculating tissue structure measurement parameters, the device comprising: The acquisition module is used to acquire computed tomography angiography (CTA) image data of the target area; wherein, the CTA image data includes multiple two-dimensional slice images; The segmentation module is used to segment the target tissue structure in the multiple two-dimensional slice images to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image; The reconstruction module is used to perform three-dimensional reconstruction based on the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image, and generate a three-dimensional geometric model of the target tissue structure. The calculation module is used to extract the geometric features of the target tissue structure from the three-dimensional geometric model, and calculate at least one type of geometric measurement parameters of the target tissue structure based on the geometric features.

[0016] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the calculation method for the tissue structure measurement parameters as described above.

[0017] This application also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the calculation method for tissue structure measurement parameters as described above.

[0018] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the calculation method for tissue structure measurement parameters as described above.

[0019] This application provides a method, apparatus, and electronic device for calculating tissue structure measurement parameters. The method involves acquiring computed tomography (CTA) angiography image data of a target region. The CTA image data includes multiple two-dimensional slice images. The target tissue structure in each of the multiple two-dimensional slice images is segmented to obtain a two-dimensional segmentation boundary for the target tissue structure in each two-dimensional slice image. Based on the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image, a three-dimensional reconstruction is performed to generate a three-dimensional geometric model of the target tissue structure. Geometric features of the target tissue structure are extracted from the three-dimensional geometric model, and at least one type of geometric measurement parameter of the target tissue structure is calculated based on these geometric features. Therefore, this application overcomes the limitation of traditional two-dimensional measurements in accurately reflecting the spatial morphology of complex tissue structures by automatically extracting features and calculating parameters from the two-dimensional segmentation results of CTA images to reconstruct a three-dimensional geometric model, thereby significantly improving the accuracy and consistency of the measurement. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating the method for calculating tissue structure measurement parameters provided in the embodiments of this application.

[0022] Figure 2 This is a schematic diagram of the structure of the calculation device for tissue structure measurement parameters provided in the embodiments of this application.

[0023] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the embodiments of this application.

[0025] The following description, in conjunction with the accompanying drawings, describes a method, apparatus, and electronic device for calculating tissue structure measurement parameters according to embodiments of this application.

[0026] Figure 1 This is a flowchart illustrating the calculation method for tissue structure measurement parameters provided in an embodiment of this application, as shown below. Figure 1 As shown, it includes the following: Step 100: Obtain computed tomography (CTA) image data of the target area; wherein the CTA image data includes multiple two-dimensional slice images.

[0027] It should be noted that this application can be widely used in the automated geometric parameter measurement of heart valves (such as mitral valve, aortic valve, tricuspid valve), blood vessels, chordae tendineae, and other anatomical structures with strip-like, sheet-like, or tubular features.

[0028] Specifically, the process begins by acquiring computed tomography (CTA) images of the target region, typically output as a series of parallel or nearly parallel two-dimensional slices. These two-dimensional slices collectively constitute the three-dimensional spatial sampling information of the target tissue structure.

[0029] Step 200: Segment the target tissue structure in the multiple two-dimensional slice images to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image.

[0030] Specifically, a target tissue structure refers to a specific anatomical entity in an image that a physician or researcher is interested in and that needs to be measured and analyzed, such as a segment of blood vessel, a leaflet, or a tendineae. For each two-dimensional slice image, the target tissue structure is segmented. Segmentation refers to distinguishing pixels or voxels belonging to the target structure from the background and other tissues in the image, obtaining its closed contour at the current slice level, i.e., the two-dimensional segmentation boundary.

[0031] Step 300: Perform three-dimensional reconstruction based on the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image to generate a three-dimensional geometric model of the target tissue structure.

[0032] Specifically, by establishing the spatial coordinate relationship between each two-dimensional slice image, the two-dimensional segmentation boundaries extracted from all slices are registered, interpolated and connected and surface reconstructed in three-dimensional space, thereby generating a three-dimensional geometric model of the target tissue structure. This three-dimensional geometric model can be a set of spatial points expressed in the form of a point cloud, or a three-dimensional surface model composed of triangular meshes, etc., which is used to digitally reproduce the real spatial morphology of the target structure in a computer.

[0033] Step 400: Extract the geometric features of the target tissue structure from the three-dimensional geometric model, and calculate at least one type of geometric measurement parameters of the target tissue structure based on the geometric features.

[0034] Specifically, geometric features that can characterize its spatial orientation, surface shape, and contour morphology are further extracted from the three-dimensional geometric model. These geometric features include, for example, the spatial curve representing the orientation of the structure's center, i.e., the center path, the normal direction perpendicular to the local surface at each point on the surface, and the boundary contour of the cross section.

[0035] Finally, based on the extracted geometric features, at least one type of geometric measurement parameter is automatically calculated. Here, geometric measurement parameters refer to numerical indicators that can quantitatively describe the size, shape, and spatial position of tissue structures, such as the true length of curved structures, the local thickness of a certain region, the major and minor axes of the contour, the spatial distance between two endpoints, or two-dimensional and three-dimensional areas, thereby providing objective and accurate quantitative basis for clinical diagnosis and preoperative planning.

[0036] The above describes the steps of the method for calculating tissue structure measurement parameters provided in the embodiments of this application. As can be seen from the above description, according to the method for calculating tissue structure measurement parameters provided in the embodiments of this application, computed tomography angiography (CTA) image data of the target region is acquired; wherein, the CTA image data includes multiple two-dimensional slice images; the target tissue structure in the multiple two-dimensional slice images is segmented to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image; three-dimensional reconstruction is performed based on the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image to generate a three-dimensional geometric model of the target tissue structure; geometric features of the target tissue structure are extracted from the three-dimensional geometric model, and at least one type of geometric measurement parameter of the target tissue structure is calculated based on the geometric features. Therefore, the embodiments of this application, by reconstructing the two-dimensional segmentation results of CTA images into a three-dimensional geometric model and automatically extracting features and calculating parameters from it, overcome the shortcomings of traditional two-dimensional measurement in failing to truly reflect the spatial morphology of complex tissue structures, thereby significantly improving the accuracy and consistency of measurement.

[0037] Based on the above embodiments, in this embodiment, step 200 segments the target tissue structure in the plurality of two-dimensional slice images to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image, including: Step 210: Preprocess each of the multiple two-dimensional slice images to obtain a preprocessed image.

[0038] Step 220: Apply a preset strategy to the preprocessed image to extract the contour, and obtain the closed contour of the target tissue structure as the two-dimensional segmentation boundary; The preset strategies include threshold segmentation, region growing, edge detection, or deep learning segmentation networks.

[0039] Specifically, in actual processing, the original CTA two-dimensional slice images often have problems such as noise interference, uneven contrast, or blurred tissue boundaries. Therefore, each two-dimensional slice image needs to be preprocessed first in order to more accurately identify the contours of the target structure. For example, grayscale normalization can be used to map the pixel values ​​of the image to a uniform range to eliminate the brightness differences between different scanning batches. Gaussian filtering or median filtering can also be used to suppress noise and smooth the image. Contrast enhancement or edge enhancement algorithms can also be used to highlight the boundary between the target tissue and the surrounding background.

[0040] After image preprocessing, preprocessed images are obtained. Then, a preset strategy is applied to these images for contour extraction, thereby obtaining the closed contour of the target tissue structure as the corresponding two-dimensional segmentation boundary in the two-dimensional slice. The preset strategy refers to a pre-selected image segmentation algorithm capable of separating the foreground target from the background, which can be flexibly selected according to different application scenarios and image characteristics. For example, thresholding segmentation sets one or more grayscale thresholds based on the difference in grayscale values ​​between the target tissue and the background in an image, dividing pixels into target and background categories; region growing first manually or automatically selects a seed point located inside the target region, and then continuously merges neighboring pixels according to a preset similarity criterion until it can no longer grow, ultimately forming a continuous target region; edge detection locates the position of the target boundary by calculating the gradient change of pixel grayscale in the image, and commonly used operators include the Canny operator and the Sobel operator; deep learning segmentation networks use pre-trained convolutional neural network models, such as U-Net and DeepLab, to perform pixel-level semantic segmentation on the input slice image, and can automatically learn and recognize the complex texture and shape features of the target tissue, especially suitable for complex situations that are difficult for traditional algorithms to handle, such as blurred boundaries or calcified regions.

[0041] Regardless of the strategy used, the final output is a closed contour curve, which defines the interface between the target tissue structure and the surrounding tissue in the current slice, and constitutes the two-dimensional segmentation boundary required for subsequent three-dimensional reconstruction.

[0042] The method for calculating tissue structure measurement parameters provided in this embodiment, by preprocessing the two-dimensional slice image and flexibly selecting various strategies for contour extraction, can effectively suppress image noise and enhance the clarity of tissue boundaries, thereby obtaining more accurate and complete two-dimensional segmentation boundaries, providing a reliable guarantee for the accuracy of subsequent three-dimensional geometric model reconstruction and parameter measurement.

[0043] Based on the above embodiments, in this embodiment, step 300 performs three-dimensional reconstruction based on the two-dimensional segmentation boundaries of the target tissue structure in each two-dimensional slice image to generate a three-dimensional geometric model of the target tissue structure, including: Step 310: Establish the spatial coordinate relationship between the multiple two-dimensional slice images.

[0044] Step 320: Based on the spatial coordinate relationship, interpolate and connect the two-dimensional segmentation boundaries of adjacent slices, and generate a three-dimensional geometric model composed of point clouds or meshes through a surface reconstruction algorithm.

[0045] Specifically, after the segmentation of the target tissue structure in all two-dimensional slice images is completed, the corresponding two-dimensional segmentation boundary has been obtained on each slice. However, these boundaries are still closed contours scattered on their respective independent planes. To restore the true spatial morphology of the target tissue structure, it is necessary to integrate and reconstruct these discrete two-dimensional contours in three-dimensional space.

[0046] Therefore, it is first necessary to establish the spatial coordinate relationship between multiple two-dimensional slice images. For example, during CTA scanning, the slice spacing, slice thickness, and translation and rotation parameters of the slice plane relative to the coordinate system of the scanning device are usually recorded. Based on these imaging parameters, the pixel coordinates in each slice can be transformed into a unified three-dimensional spatial coordinate system, so that the originally isolated two-dimensional contour points can obtain real spatial coordinates, thereby providing a geometric basis for subsequent contour connections.

[0047] After establishing a unified spatial coordinate relationship, interpolation is performed to connect the two-dimensional segmentation boundaries of adjacent slices. That is, between the closed contours of two adjacent slices, an intermediate transition contour is generated through linear interpolation or spline interpolation according to a certain spatial sampling step size. This makes the slice data with large gaps between layers continuous, thus avoiding the step-like interlayer breakage phenomenon on the surface of the reconstructed model. Subsequently, a three-dimensional geometric model composed of point clouds or meshes is generated by surface reconstruction algorithms. For example, the moving cube algorithm can be used to extract isosurfaces from voxel data. The final output three-dimensional geometric model can be a set of spatial discrete points expressed in the form of point clouds, or a mesh model composed of triangular or polygonal facets. This model can digitally and intuitively present the three-dimensional morphology of the target tissue structure in a complete and intuitive way in the computer, including its curvature, spatial torsion, and the actual dimensions of each cross section. This provides an accurate spatial carrier for the subsequent automatic measurement of geometric parameters such as center path extraction, thickness analysis, and area calculation.

[0048] The method for calculating tissue structure measurement parameters provided in this embodiment can effectively eliminate the interlayer step effect by establishing spatial coordinate relationships between slices and interpolating and connecting adjacent contours and reconstructing surfaces, thereby generating a continuous, smooth, and realistically reflective three-dimensional geometric model of the target structure's spatial morphology, thus providing a reliable spatial carrier for the subsequent accurate automatic measurement of geometric parameters.

[0049] Based on the above embodiments, in this embodiment, the geometric features include at least one of the center path, normal direction, and boundary profile; the geometric measurement parameters include at least one of the length, thickness, major axis, minor axis, endpoint distance, two-dimensional area, and three-dimensional surface area.

[0050] In one embodiment, extracting the geometric features of the target tissue structure from the three-dimensional geometric model and calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: The center path is extracted from the three-dimensional geometric model using a skeleton extraction algorithm; By performing discrete curve integration on the central path, the spatial distances between adjacent points on the central path are calculated and accumulated to obtain the length of the target tissue structure.

[0051] Specifically, the central path is first extracted from the 3D geometric model using a skeleton extraction algorithm. For example, a thinning algorithm based on distance transformation or a mid-axis transformation algorithm based on Voronoi diagrams can be used to peel away external voxels or mesh vertices layer by layer from the 3D surface model of the target tissue, ultimately retaining a spatial curve located inside the structure and extending along its major axis—the central path. After obtaining this central path, discrete curve integration is performed. Discrete curve integration involves sampling the continuous central path curve into a series of sequentially adjacent spatial discrete points at a certain step size, calculating the Euclidean distance between adjacent points segment by segment, and then accumulating these segmented distances to obtain the total arc length along the path. Since this arc length strictly follows the actual bending trajectory of the central path, the obtained value can accurately reflect the actual length of the target tissue structure, effectively avoiding the underestimation of length caused by tissue bending or torsion, and significantly improving the accuracy of length measurement for curved structures.

[0052] In another embodiment, calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: Multiple test points are selected on the surface of the three-dimensional geometric model, and the corresponding normal direction is determined for each of the multiple test points. Along the normal direction at the point to be measured, find the intersection point of the normal direction and the other side surface of the three-dimensional geometric model, and take it as the corresponding point; Calculate the spatial distance between the point to be measured and the corresponding point, and use it as the local thickness at the point to be measured; The thickness distribution of the target tissue structure is obtained based on the local thickness of the multiple test points.

[0053] Specifically, multiple test points are selected on the surface of the generated 3D geometric model. These test points can be user-specified locations of interest or automatically determined sampling points based on uniform sampling or curvature features. For each test point, its normal direction is first determined based on the geometry of the local area on the surface where the point is located. The normal direction is perpendicular to the model surface at that point and points outward or inward, and can be calculated using methods such as weighted averaging of the normal vectors of the triangular facets in the neighborhood of that point. Subsequently, a virtual spatial ray is drawn along the normal direction at the test point, penetrating the 3D geometric model until it hits the inner wall surface or the opposite surface on the other side of the model. The intersection of the ray and the opposite surface is the corresponding point, which forms a spatial correspondence with the test point based on normal direction matching. The spatial distance between the test point and the corresponding point is calculated, and the resulting distance value is the local thickness at the test point, reflecting the wall thickness of the tissue along the normal direction at the current location. By performing the above steps on each of the multiple test points, a series of local thickness values ​​distributed on the model surface can be obtained. By mapping and summarizing these local thickness values ​​on the model surface, the thickness distribution of the target tissue structure can be obtained. For example, the thickness variation and spatial trend of the tissue wall can be intuitively displayed in the form of a color-coded thickness cloud map, thereby providing a comprehensive and objective quantitative basis for assessing tissue lesion areas and identifying abnormally thickened or thinned areas.

[0054] In yet another embodiment, calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: Ellipse fitting or principal direction analysis is performed on the boundary contour to determine the major and minor axes of the target tissue structure. The length of the major axis and the length of the minor axis are respectively used as the major axis and minor axis of the target tissue structure.

[0055] Specifically, using the extracted boundary contours—that is, the closed curves of the outer edges of the target tissue structure in a certain cross section or local region of a 3D model—ellipse fitting or principal direction analysis is performed. Ellipse fitting refers to finding a standard ellipse equation that best matches the current boundary contour using optimization algorithms such as least squares, minimizing the sum of the squared distances from all contour points to the edge of the ellipse. Principal direction analysis involves calculating the covariance matrix of the boundary contour point set and solving for its eigenvalues ​​and eigenvectors, using the direction of the eigenvector corresponding to the largest eigenvalue as the principal direction of the data distribution. After the above processing, the major and minor axes of the target tissue structure can be determined. The major axis corresponds to the principal axis direction of the fitted ellipse or the direction of the largest eigenvector in the principal direction analysis, while the minor axis corresponds to the secondary axis direction orthogonal to it. Finally, the lengths of the major and minor axes are output as the major and minor axes of the target tissue structure, respectively, providing objective quantitative evidence for judging whether the valve annulus is dilated or the lumen is deformed.

[0056] In another embodiment, calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: The endpoint positions of the target tissue structure are identified based on the curvature changes or endpoint features of the boundary contour. Calculate the spatial distance between the identified endpoints as the endpoint distance of the target organizational structure.

[0057] Specifically, the extracted boundary contours are used for geometric analysis to automatically identify the endpoint positions of the target tissue structure. On the one hand, endpoints can be located by calculating the curvature changes of each point on the boundary contour. On the other hand, for contours with naturally open ends, endpoint positions can also be directly identified through boundary endpoint features. Boundary endpoint features refer to the geometric characteristics of closed or incompletely closed contours at their endpoints, such as the contour line being interrupted at that point, or the number of pixel connections in the neighborhood abruptly changing to only one adjacent point. After successfully identifying at least two endpoint positions, the spatial distance between the identified endpoints is calculated. If the endpoints are defined on a two-dimensional contour plane, the distance is a Euclidean straight line distance; if the endpoint positions are mapped to three-dimensional space via a three-dimensional model, the three-dimensional Euclidean distance is calculated. The result is the endpoint distance of the target tissue structure, thus providing an objective and accurate quantitative basis for evaluating chordae tendineae length, leaflet free edge span, etc.

[0058] In yet another embodiment, calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: The two-dimensional area of ​​the target tissue structure is obtained by integrating the closed contour formed by the two-dimensional segmentation boundary. Based on the mesh representation of the three-dimensional geometric model, the areas of the triangular facets constituting the surface are accumulated to obtain the three-dimensional surface area of ​​the target tissue structure.

[0059] Specifically, when calculating the two-dimensional area, integration is performed using the closed contour formed by the two-dimensional segmentation boundary. The two-dimensional segmentation boundary is the outer contour of the target tissue extracted in the aforementioned segmentation step within a single slice image. This contour is a closed curve connected end to end. By counting the pixel region enclosed by this closed contour or converting the contour coordinates into area integral values ​​using Green's formula, the two-dimensional area of ​​the cross section can be obtained. This method is suitable for evaluating planar area measurements such as blood vessel cross-sectional area and valve orifice opening area. When calculating the three-dimensional surface area, processing is performed based on the mesh representation of the three-dimensional geometric model. The mesh representation is a digital expression of the three-dimensional geometric model that approximates the real surface by piecing together numerous tiny triangular facets. Each triangular facet is uniquely defined by the spatial coordinates of its three vertices. The area of ​​each triangular facet is calculated sequentially using Heron's formula or the cross product method. Then, the areas of all triangular facets constituting the model surface are summed to obtain the three-dimensional surface area of ​​the target tissue structure. This value can accurately reflect the actual surface coverage of the target structure in three-dimensional space, providing a reliable quantitative basis for evaluating complex surface measurements such as valve leaflet area and tissue defect extent.

[0060] The method for calculating tissue structure measurement parameters provided in this embodiment extracts multiple geometric features such as the center path, normal direction, and boundary contour, and realizes automated and accurate calculation of length, thickness distribution, major and minor axes, endpoint distance, and two-dimensional and three-dimensional area, thereby comprehensively covering the accurate measurement needs of multiple types of geometric parameters of complex tissue structures under a unified framework.

[0061] The following describes the calculation device for tissue structure measurement parameters provided in the embodiments of this application. The calculation device and calculation method for tissue structure measurement parameters described below can be referred to in correspondence with each other.

[0062] Figure 2 This is a schematic diagram of the structure of the calculation device for tissue structure measurement parameters provided in the embodiments of this application, as shown below. Figure 2 As shown, the tissue structure measurement parameter calculation device provided in this application embodiment includes: The acquisition module 201 is used to acquire computed tomography angiography (CTA) image data of the target area; wherein, the CTA image data includes multiple two-dimensional slice images; The segmentation module 202 is used to segment the target tissue structure in the plurality of two-dimensional slice images to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image; Reconstruction module 203 is used to perform three-dimensional reconstruction based on the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image, and generate a three-dimensional geometric model of the target tissue structure; The calculation module 204 is used to extract the geometric features of the target tissue structure from the three-dimensional geometric model, and calculate at least one type of geometric measurement parameters of the target tissue structure based on the geometric features.

[0063] The tissue structure measurement parameter calculation device provided in this application embodiment acquires computed tomography angiography (CTA) image data of a target region; wherein the CTA image data includes multiple two-dimensional slice images; the target tissue structure in the multiple two-dimensional slice images is segmented to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image; three-dimensional reconstruction is performed based on the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image to generate a three-dimensional geometric model of the target tissue structure; geometric features of the target tissue structure are extracted from the three-dimensional geometric model, and at least one type of geometric measurement parameter of the target tissue structure is calculated based on the geometric features. Therefore, this application embodiment overcomes the deficiency of traditional two-dimensional measurement in failing to accurately reflect the spatial morphology of complex tissue structures by reconstructing the two-dimensional segmentation results of CTA images into a three-dimensional geometric model and automatically extracting features and calculating parameters from it, thereby significantly improving the accuracy and consistency of measurement.

[0064] Based on the above embodiments, in this embodiment, the segmentation module 202 is specifically used for: Each of the plurality of two-dimensional slice images is preprocessed to obtain a preprocessed image; A preset strategy is applied to the preprocessed image to extract the contour, and the closed contour of the target tissue structure is used as the two-dimensional segmentation boundary. The preset strategies include threshold segmentation, region growing, edge detection, or deep learning segmentation networks.

[0065] Based on the above embodiments, in this embodiment, the reconstruction module 203 is specifically used for: Establish the spatial coordinate relationship between the multiple two-dimensional slice images; Based on the spatial coordinate relationship, the two-dimensional segmentation boundaries of adjacent slices are interpolated and connected, and a three-dimensional geometric model composed of point clouds or meshes is generated by the surface reconstruction algorithm.

[0066] Based on the above embodiments, in this embodiment... The geometric features include at least one of the center path, normal direction, and boundary profile; The geometric measurement parameters include at least one of length, thickness, major diameter, minor diameter, endpoint distance, two-dimensional area, and three-dimensional surface area.

[0067] Based on the above embodiments, in this embodiment, the calculation module 204 is specifically used for: The center path is extracted from the three-dimensional geometric model using a skeleton extraction algorithm; By performing discrete curve integration on the central path, the spatial distances between adjacent points on the central path are calculated and accumulated to obtain the length of the target tissue structure.

[0068] Based on the above embodiments, in this embodiment, the calculation module 204 is further configured to: Multiple test points are selected on the surface of the three-dimensional geometric model, and the corresponding normal direction is determined for each of the multiple test points. Along the normal direction at the point to be measured, find the intersection point of the normal direction and the other side surface of the three-dimensional geometric model, and take it as the corresponding point; Calculate the spatial distance between the point to be measured and the corresponding point, and use it as the local thickness at the point to be measured; The thickness distribution of the target tissue structure is obtained based on the local thickness of the multiple test points.

[0069] Based on the above embodiments, in this embodiment, the calculation module 204 is further configured to: Ellipse fitting or principal direction analysis is performed on the boundary contour to determine the major and minor axes of the target tissue structure. The length of the major axis and the length of the minor axis are respectively used as the major axis and minor axis of the target tissue structure.

[0070] Based on the above embodiments, in this embodiment, the calculation module 204 is further configured to: The endpoint positions of the target tissue structure are identified based on the curvature changes or endpoint features of the boundary contour. Calculate the spatial distance between the identified endpoints as the endpoint distance of the target organizational structure.

[0071] Based on the above embodiments, in this embodiment, the calculation module 204 is further configured to: The two-dimensional area of ​​the target tissue structure is obtained by integrating the closed contour formed by the two-dimensional segmentation boundary. Based on the mesh representation of the three-dimensional geometric model, the areas of the triangular facets constituting the surface are accumulated to obtain the three-dimensional surface area of ​​the target tissue structure.

[0072] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3As shown, the electronic device can be a robot or other electronic device. This electronic device may include: a processor 310, a communications interface 320, a memory 330, and a communication bus 340. The processor 310, communications interface 320, and memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions from the memory 330 to execute methods for calculating organizational structure measurement parameters, including: Acquire computed tomography (CTA) image data of the target area; wherein the CTA image data includes multiple two-dimensional slice images; The target tissue structure in the plurality of two-dimensional slice images is segmented to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image; Based on the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image, a three-dimensional reconstruction is performed to generate a three-dimensional geometric model of the target tissue structure. Geometric features of the target tissue structure are extracted from the three-dimensional geometric model, and at least one type of geometric measurement parameters of the target tissue structure are calculated based on the geometric features.

[0073] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application embodiment, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in at least one embodiment of this application embodiment. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0074] On the other hand, embodiments of this application also provide a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the calculation methods for tissue structure measurement parameters provided by the above methods, including: Acquire computed tomography (CTA) image data of the target area; wherein the CTA image data includes multiple two-dimensional slice images; The target tissue structure in the plurality of two-dimensional slice images is segmented to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image; Based on the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image, a three-dimensional reconstruction is performed to generate a three-dimensional geometric model of the target tissue structure. Geometric features of the target tissue structure are extracted from the three-dimensional geometric model, and at least one type of geometric measurement parameters of the target tissue structure are calculated based on the geometric features.

[0075] In another aspect, embodiments of this application also provide a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a method for calculating tissue structure measurement parameters provided by the methods described above, including: Acquire computed tomography (CTA) image data of the target area; wherein the CTA image data includes multiple two-dimensional slice images; The target tissue structure in the plurality of two-dimensional slice images is segmented to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image; Based on the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image, a three-dimensional reconstruction is performed to generate a three-dimensional geometric model of the target tissue structure. Geometric features of the target tissue structure are extracted from the three-dimensional geometric model, and at least one type of geometric measurement parameters of the target tissue structure are calculated based on the geometric features.

[0076] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0077] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of this application, and are not intended to limit them; although the embodiments of this application have been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for calculating tissue structure measurement parameters, characterized in that, include: Acquire computed tomography (CTA) image data of the target area; wherein the CTA image data includes multiple two-dimensional slice images; The target tissue structure in the plurality of two-dimensional slice images is segmented to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image; Based on the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image, a three-dimensional reconstruction is performed to generate a three-dimensional geometric model of the target tissue structure. Geometric features of the target tissue structure are extracted from the three-dimensional geometric model, and at least one type of geometric measurement parameters of the target tissue structure are calculated based on the geometric features.

2. The method for calculating tissue structure measurement parameters according to claim 1, characterized in that, The step of segmenting the target tissue structure in the plurality of two-dimensional slice images to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image includes: Each of the plurality of two-dimensional slice images is preprocessed to obtain a preprocessed image; A preset strategy is applied to the preprocessed image to extract the contour, and the closed contour of the target tissue structure is used as the two-dimensional segmentation boundary. The preset strategies include threshold segmentation, region growing, edge detection, or deep learning segmentation networks.

3. The method for calculating tissue structure measurement parameters according to claim 1 or 2, characterized in that, The step of performing three-dimensional reconstruction based on the two-dimensional segmentation boundaries of the target tissue structure in each two-dimensional slice image to generate a three-dimensional geometric model of the target tissue structure includes: Establish the spatial coordinate relationship between the multiple two-dimensional slice images; Based on the spatial coordinate relationship, the two-dimensional segmentation boundaries of adjacent slices are interpolated and connected, and a three-dimensional geometric model composed of point clouds or meshes is generated by the surface reconstruction algorithm.

4. The method for calculating tissue structure measurement parameters according to any one of claims 1 to 3, characterized in that, The geometric features include at least one of the center path, normal direction, and boundary profile; The geometric measurement parameters include at least one of length, thickness, major diameter, minor diameter, endpoint distance, two-dimensional area, and three-dimensional surface area.

5. The method for calculating tissue structure measurement parameters according to claim 4, characterized in that, The step of extracting geometric features of the target tissue structure from the three-dimensional geometric model and calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: The center path is extracted from the three-dimensional geometric model using a skeleton extraction algorithm; The length of the target tissue structure is obtained by performing discrete curve integration on the central path, calculating and summing the spatial distances between adjacent points on the central path.

6. The method for calculating tissue structure measurement parameters according to claim 4 or 5, characterized in that, The step of calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: Multiple test points are selected on the surface of the three-dimensional geometric model, and the corresponding normal direction is determined for each of the multiple test points. Along the normal direction at the point to be measured, find the intersection point of the normal direction and the other side surface of the three-dimensional geometric model, and take it as the corresponding point; Calculate the spatial distance between the point to be measured and the corresponding point, and use it as the local thickness at the point to be measured; The thickness distribution of the target tissue structure is obtained based on the local thickness of the multiple test points.

7. The method for calculating tissue structure measurement parameters according to any one of claims 4 to 6, characterized in that, The step of calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: Ellipse fitting or principal direction analysis is performed on the boundary contour to determine the major and minor axes of the target tissue structure. The length of the major axis and the length of the minor axis are respectively used as the major axis and minor axis of the target tissue structure.

8. The method for calculating tissue structure measurement parameters according to any one of claims 4 to 7, characterized in that, The step of calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: The endpoint positions of the target tissue structure are identified based on the curvature changes or endpoint features of the boundary contour. Calculate the spatial distance between the identified endpoints as the endpoint distance of the target organizational structure.

9. The method for calculating tissue structure measurement parameters according to any one of claims 4 to 8, characterized in that, The step of calculating at least one type of geometric measurement parameters of the target tissue structure based on the geometric features includes: The two-dimensional area of ​​the target tissue structure is obtained by integrating the closed contour formed by the two-dimensional segmentation boundary. Based on the mesh representation of the three-dimensional geometric model, the areas of the triangular facets constituting the surface are accumulated to obtain the three-dimensional surface area of ​​the target tissue structure.

10. A calculation device for tissue structure measurement parameters, characterized in that, include: The acquisition module is used to acquire computed tomography angiography (CTA) image data of the target area; wherein, the CTA image data includes multiple two-dimensional slice images; The segmentation module is used to segment the target tissue structure in the multiple two-dimensional slice images to obtain the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image; The reconstruction module is used to perform three-dimensional reconstruction based on the two-dimensional segmentation boundary of the target tissue structure in each two-dimensional slice image, and generate a three-dimensional geometric model of the target tissue structure. The calculation module is used to extract the geometric features of the target tissue structure from the three-dimensional geometric model, and calculate at least one type of geometric measurement parameters of the target tissue structure based on the geometric features.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for calculating the tissue structure measurement parameters as described in any one of claims 1 to 9.

12. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method for calculating the tissue structure measurement parameters as described in any one of claims 1 to 9.

13. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method for calculating the tissue structure measurement parameters as described in any one of claims 1 to 9.

Images