Mechanical analysis method, device, equipment and storage medium for heart valve region

Abstract

The present invention relates to the field of medical technology, and specifically to a mechanical analysis method, apparatus, device, and storage medium for a heart valve region, the method comprising: acquiring a plurality of medical image data corresponding to different moments in a cardiac cycle containing a heart valve region; performing image segmentation processing on each of the medical image data to obtain a plurality of image segmentation results corresponding to the heart valve region; generating a dynamic model of the heart valve region based on the plurality of image segmentation results; and determining the mechanical distribution information of the heart valve region using a non-steady-state simulation method based on the dynamic model. The mechanical analysis method for a heart valve region provided by the present invention can obtain mechanical distribution information that is more in line with actual conditions by generating a dynamic model of the heart valve region and performing non-steady-state numerical simulation based on the dynamic model, thereby improving the accuracy of the mechanical analysis results.

Description

Technical Field

[0001] The present invention relates to the field of medical technology, and in particular to a mechanical analysis method, device, equipment and storage medium for a heart valve region. Background Art

[0002] Cardiovascular disease is currently the disease with the highest morbidity and mortality. Among them, heart valve disease is a cardiovascular disease that seriously threatens life and health. Every year, more than 200,000 patients need valve surgery, ranking first among adult heart surgeries.

[0003] Heart valves are valves located between the atria and ventricles, or between the ventricles and arteries. These include the mitral valve, tricuspid valve, pulmonary valve, and aortic valve. The valves open and close via a pressure differential across the valve leaflets, ensuring unidirectional blood flow and minimizing backflow. The aortic valve, located between the left ventricle and the aorta, where pressure loads are greater, is particularly susceptible to aortic valve disease. Aortic valve disease often results in abnormal opening and closing of the aortic valve, leading to problems such as insufficient cardiac output and impacting the body's normal physiological state.

[0004] Aortic stenosis (AS) is the most common aortic valve disease in the elderly. Its pathophysiology is degenerative and calcific. Symptomatic patients can undergo minimally invasive transcatheter aortic valve replacement (TAVR). TAVR utilizes a minimally invasive cardiac catheterization technique to replace an artificial heart valve. TAVR offers advantages such as no thoracotomy, low risk, ease of operation, minimal trauma, rapid recovery, and few complications. It is particularly suitable for patients with contraindications to surgery or those at high risk for surgery.

[0005] The current preoperative planning schemes for TAVR surgery are mostly based on the doctor's experience, and the success rate and feasibility of the surgery depend largely on the doctor's technical level. In recent years, thanks to the development of numerical technology and computing power, researchers have also conducted a lot of research on the aortic valve to optimize the preoperative planning schemes for TAVR surgery and guide the surgical process. However, the existing technologies (such as the patent document with publication number CN110993111A) mostly perform steady-state structural analysis of the aortic valve based on a single morphological model. Although some valuable conclusions can be drawn, since the heart is always in a beating state, the results obtained based on the single structural analysis are still far from the actual situation. Therefore, there is an urgent need to provide a new technical solution based on the mechanical analysis method of the heart valve region to accurately determine the mechanical distribution information of the heart valve region, thereby meeting the actual corresponding technical indicator requirements. Summary of the Invention

[0006] In view of the above-mentioned problems in the prior art, the object of the present invention is to provide a mechanical analysis method, device, equipment and storage medium of a heart valve region, which can more accurately determine the mechanical distribution information of the heart valve region.

[0007] In order to solve the above problems, the present invention provides a mechanical analysis method for a heart valve region, comprising:

[0008] S210: Acquire a plurality of medical image data including a heart valve region corresponding to different moments in a cardiac cycle;

[0009] S220: performing image segmentation processing on each of the medical image data to obtain a plurality of image segmentation results corresponding to the heart valve region;

[0010] S230: generating a dynamic model of the heart valve region according to the multiple image segmentation results;

[0011] S240: Based on the dynamic model, determine the mechanical distribution information of the heart valve region using a non-steady-state simulation method.

[0012] Furthermore, the step S220 includes:

[0013] S221: extracting first image data of the heart valve region from each medical image data respectively;

[0014] S222: Perform image segmentation processing on the first image data of the heart valve region to obtain an image segmentation result corresponding to the heart valve region.

[0015] Furthermore, the step S222 includes:

[0016] S410: performing image segmentation processing on the first image data to obtain a first mask image of the heart valve region;

[0017] S420: generating a surface mesh of the first mask image using a marching cubes algorithm, where the surface mesh includes a plurality of mesh vertices;

[0018] S430: Extracting feature data corresponding to each mesh vertex from the medical image data;

[0019] S440: Constructing a graph model based on the surface mesh and feature data corresponding to each mesh vertex;

[0020] S450: Solve the graph model using a graph cut algorithm to obtain a multi-classification result of the image grid corresponding to the heart valve region.

[0021] Furthermore, the step S430 includes:

[0022] S431: Based on a preset feature extraction operator or deep learning network, extract texture information within a preset range of each mesh vertex from the medical image data as feature data corresponding to each mesh vertex.

[0023] Optionally, the step S431 includes:

[0024] performing a morphological operation on the first mask image using a morphological algorithm to obtain a second mask image;

[0025] applying the second mask image to the medical image data to obtain second image data of the heart valve region;

[0026] determining a target image region in the medical image data based on the second image data;

[0027] Based on a preset feature extraction operator or a deep learning network, texture information in the projection direction corresponding to each of the mesh vertices is extracted from the target image area as feature data corresponding to each of the mesh vertices.

[0028] Furthermore, the image segmentation result includes surface grid structures corresponding to a plurality of heart parts, and the plurality of heart parts include at least one valve and corresponding related parts;

[0029] The step S230 includes:

[0030] for each heart region, determining a plurality of surface mesh structures corresponding to the heart region, and generating a dynamic model of the heart region according to the plurality of surface mesh structures;

[0031] A dynamic model of the heart valve region is generated based on the dynamic models of the respective heart regions.

[0032] Furthermore, generating the dynamic model of the heart region includes:

[0033] Optimizing and reconstructing each of the surface grid structures to form a corresponding surface structure;

[0034] Respectively performing volume reconstruction and spatial meshing on the area surrounded by each surface structure to form a corresponding volume structure;

[0035] A dynamic model of the heart region is generated based on the volume structures corresponding to the surface mesh structures.

[0036] Furthermore, the step S240 includes:

[0037] Performing solid domain numerical simulation based on the dynamic model to obtain structural mechanical parameters of various positions in the heart valve region; and / or,

[0038] obtaining a blood flow velocity in the heart valve region as a boundary condition;

[0039] Fluid domain numerical simulation is performed based on the dynamic model and the blood flow velocity to obtain hemodynamic parameters of the heart valve region.

[0040] Furthermore, the method further comprises:

[0041] S250: determining a location with abnormal mechanical distribution in the heart valve region according to the mechanical distribution information;

[0042] S260: Prompt the location where the mechanical distribution is abnormal.

[0043] Another aspect of the present invention provides a mechanical analysis device for a heart valve region, comprising:

[0044] A medical image data acquisition module, configured to acquire a plurality of medical image data including a heart valve region corresponding to different moments in a cardiac cycle;

[0045] An image segmentation module, configured to perform image segmentation processing on each of the medical image data to obtain a plurality of image segmentation results corresponding to the heart valve region;

[0046] a dynamic model generating module, configured to generate a dynamic model of the heart valve region according to the plurality of image segmentation results;

[0047] The mechanical distribution information determination module is used to determine the mechanical distribution information of the heart valve region based on the dynamic model using a non-steady-state simulation method.

[0048] On the other hand, the present invention provides an electronic device comprising a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the mechanical analysis method of the heart valve area as described above.

[0049] On the other hand, the present invention provides a computer-readable storage medium, which stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the mechanical analysis method of the heart valve area as described above.

[0050] Due to the above technical solution, the present invention has the following beneficial effects:

[0051] According to the mechanical analysis method of the heart valve area in an embodiment of the present invention, multiple medical image data within a cardiac cycle are segmented to obtain multiple image segmentation results corresponding to the heart valve area, and a dynamic model of the heart valve area is generated based on the multiple image segmentation results. By performing non-steady-state numerical simulation based on the dynamic model, mechanical distribution information that is more in line with the actual situation can be obtained, which can truly reflect the mechanical changes of the heart valve area during movement, thereby improving the accuracy of the mechanical analysis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] To more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present invention, and those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0053] Figure 1 This is a schematic diagram of an implementation environment provided by an embodiment of the present invention;

[0054] Figure 2 This is a flow chart of a mechanical analysis method for a heart valve region provided by one embodiment of the present invention;

[0055] Figure 3 is a schematic diagram of first image data of an aortic valve region provided by one embodiment of the present invention;

[0056] Figure 4 is a flowchart of an image segmentation process provided by one embodiment of the present invention;

[0057] Figure 5 is a schematic diagram of the open and closed states of a normal aortic valve provided by one embodiment of the present invention;

[0058] Figure 6 is a schematic diagram of the open state of the aortic valve in a CT angiography image provided by one embodiment of the present invention;

[0059] Figure 7 is a flow chart of a mechanical analysis method for a heart valve region provided by another embodiment of the present invention;

[0060] Figure 8 1 is a schematic structural diagram of a mechanical analysis device for a heart valve region provided by one embodiment of the present invention;

[0061] Figure 9 It is a structural diagram of an electronic device provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0063] It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product, or device comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0064] Since the tissue characteristics of the heart valve will change in a pathological state, which will cause the state of the heart valve to change during the whole cardiac cycle, the morphological changes of the heart valve regional structure can be obtained by segmenting the complete cardiac cycle CT image, and then the mechanical changes during the movement can be obtained based on the numerical simulation method, and finally the abnormal part can be located, which provides a basis for the selection of valve prosthesis in TAVR surgery, thereby optimizing the preoperative planning of TAVR surgery. In order to make the purpose, technical solutions and advantages disclosed in the embodiments of the present invention clearer, the embodiments of the present invention are further described in detail in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the embodiments of the present invention and are not used to limit the embodiments of the present invention.

[0065] First, the following concepts are explained in the embodiments of the present invention:

[0066] Convolutional Neural Networks (CNN): CNN is a type of feedforward neural network that includes convolution calculations and has a deep structure. It is one of the representative algorithms of deep learning.

[0067] Graph Cut Algorithm: Graph Cut algorithms are commonly used in computer vision to effectively solve various computer vision problems, such as image smoothing, stereo mapping, and image segmentation. These methods relate the image segmentation problem to the minimum cut problem on a graph. In many similar computer vision problems, the minimum energy solution corresponds to the maximum a posteriori estimate of the solution.

[0068] Fluid mechanics simulation: Fluid mechanics simulation is an interdisciplinary subject between mathematics, fluid mechanics and computers. Its main research content is to solve the governing equations of fluid mechanics through computers and numerical methods, and to simulate and analyze fluid mechanics problems.

[0069] Reference Manual Figure 1 , which shows a schematic diagram of an implementation environment provided by an embodiment of the present invention. Figure 1 As shown, the implementation environment may include at least one medical scanning device 110 and a computer device 120. The computer device 120 and each medical scanning device 110 may be directly or indirectly connected via wired or wireless communication, which is not limited in this embodiment of the present invention.

[0070] Among them, the computer device 120 can be but is not limited to various servers, personal computers, laptops, smart phones, tablets and portable wearable devices. The server can be an independent server or a server cluster or distributed system composed of multiple servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), as well as big data and artificial intelligence platforms.

[0071] In an embodiment of the present invention, the medical scanning device 110 can use imaging technologies such as CT imaging and magnetic resonance imaging (MRI) to image the patient's heart region, thereby acquiring multiple medical image data, including the heart valve region, during a complete cardiac cycle. The computer device 120 can acquire the medical image data acquired by the medical scanning device 110 and, using the methods provided in an embodiment of the present invention, determine the mechanical distribution information of the heart valve region, ultimately locating areas with abnormal mechanical distribution, thereby providing a basis for selecting valve prostheses during TAVR surgery and optimizing preoperative planning for TAVR surgery.

[0072] It should be noted that Figure 1 It is just an example. Those skilled in the art will understand that although Figure 1Only one medical scanning device 110 is shown in the figure, but this does not constitute a limitation on the embodiment of the present invention. Any number of medical scanning devices 110 may be provided according to actual needs.

[0073] Reference Manual Figure 2 , which shows the process of the mechanical analysis method of the heart valve region provided by an embodiment of the present invention, which can be applied to Figure 1 In the computer device 120, specifically Figure 2 As shown, the method may include the following steps:

[0074] S210: Acquire a plurality of medical image data including a heart valve region corresponding to different moments in a cardiac cycle.

[0075] In embodiments of the present invention, the heart valve region may be a heart region including a particular heart valve experiencing an abnormality and its associated parts, for example, a heart region including the mitral valve, tricuspid valve, aortic valve, and / or pulmonary valve and their associated parts. Specifically, the heart valve region may be a heart region including the aortic valve and aorta, or a heart region including the pulmonary valve and pulmonary artery, and so on.

[0076] In an embodiment of the present invention, the plurality of medical image data may be a plurality of medical image data corresponding to different moments selected from medical image data within a cardiac cycle. For example, the plurality of medical image data may be a plurality of medical image data selected with equal imaging time intervals. Specifically, the medical image data may be CT angiography image data, MRI image data, or the like.

[0077] In an embodiment of the present invention, the source of the multiple medical image data can be directly imported related data, or obtained from other resource libraries through real-time configuration connections, or obtained from a stored image database after searching based on the user's name and other information. The embodiment of the present invention does not impose any restrictions on this.

[0078] For example, a patient can be subjected to an ECG-gated scan using a CT device to obtain multiple CT angiographic image data for a complete cardiac cycle, generally represented as data from 0% to 100% of the RR interval (the period between two adjacent R waves on an electrocardiogram). Specifically, if the image acquisition time interval is set to 10% of the RR interval, CT angiographic image data from 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, and 90% of the RR interval can be obtained as multiple medical image data within a complete cardiac cycle.

[0079] S220: Perform image segmentation processing on each of the medical image data to obtain multiple image segmentation results corresponding to the heart valve region.

[0080] In an embodiment of the present invention, multiple image segmentation results corresponding to the heart valve region can be obtained by performing image segmentation processing on the multiple medical image data, and a dynamic model of the valve region can be generated based on the multiple image segmentation results. The multiple image segmentation results corresponding to the heart valve region each include segmentation results for multiple heart regions within the heart valve region, where the multiple heart regions include at least one valve and corresponding related regions.

[0081] In a possible embodiment, performing image segmentation processing on each of the medical image data to obtain multiple image segmentation results corresponding to the heart valve region (step S220) may include:

[0082] S221: extracting first image data of the heart valve region from each medical image data respectively;

[0083] S222: Perform image segmentation processing on the first image data of the heart valve region to obtain an image segmentation result corresponding to the heart valve region.

[0084] Specifically, in order to reduce the computational complexity of image segmentation, the first image data of the heart valve region may be extracted from various medical image data, and image segmentation processing may be performed only on the first image data of the heart valve region.

[0085] In practical applications, the whole heart atlas model can be used to perform template matching on medical image data to extract the first image data of the heart valve region, and also to determine information such as the valve position in the heart valve region. For example, Figure 3 As shown, the whole heart atlas model can be used to perform template matching on medical image data to obtain image data of the aorta and left ventricular cavity, and the approximate position of the aortic valve can be determined based on the junction of the aorta and left ventricular cavity. Image data containing the aorta and left ventricular cavity can be obtained as the first image data of the aortic valve region, or image data containing only the aorta and aortic valve can be obtained as the first image data.

[0086] In one possible embodiment, the Figure 4 The performing image segmentation processing on the first image data of the heart valve region to obtain an image segmentation result corresponding to the heart valve region (step S222) may include:

[0087] S410: Perform image segmentation processing on the first image data to obtain a first mask image of the heart valve region.

[0088] In an embodiment of the present invention, a more sophisticated traditional algorithm such as template matching or region growing can be used to perform image segmentation processing on the first image data to obtain a first mask image of the heart valve region. The first mask image is a binary image, where the heart valve region (i.e., the segmented foreground) corresponds to a value of 1, and the rest of the region (i.e., the background) corresponds to a value of 0.

[0089] S420: Generate a surface mesh of the first mask image using a marching cubes algorithm, where the surface mesh includes a plurality of mesh vertices.

[0090] In an embodiment of the present invention, a corresponding surface mesh can be generated based on the first mask image using the marching cubes algorithm. The specific details of the marching cubes algorithm can be referenced in the prior art and will not be further described in detail in the present embodiment. It will be appreciated that when the marching cubes algorithm is used to generate the surface mesh, a higher density of mesh vertices than pixel points can be generated, resulting in a higher resolution of the generated surface mesh than the resolution of the medical image. This can effectively preserve detailed information about the heart valves and improve image segmentation accuracy.

[0091] In the embodiment of the present invention, for different heart valves, the surface grid may include different structures. For example, for the aortic valve, refer to the attached manual. Figure 5 and Figure 6 As shown, under normal circumstances, the aortic valve includes three independent aortic valves. When the aortic valve is open, a cavity structure appears in the middle of the three aortic valves. When the valve is closed, there is no such cavity structure. The first image data of the aortic valve area extracted from the medical image data includes at least the aortic valve and the aorta. The surface mesh generated by the marching cubes algorithm can include three (one or two in the pathological state) independent aortic valve structures, the aortic structure, and the cavity structure presented when the valve is open. For another example, for the mitral valve, the first image data of the mitral valve area extracted from the medical image data includes at least the mitral valve and the left ventricle. The surface mesh generated by the marching cubes algorithm can include two (one in the pathological state) independent mitral valve structures, the left ventricle structure, and the cavity structure presented when the valve is open.

[0092] S430: Extracting feature data corresponding to each mesh vertex from the medical image data.

[0093] In an embodiment of the present invention, local texture information of each mesh vertex can be extracted from the medical image data as feature data corresponding to the mesh vertex, so that each mesh vertex on the surface mesh includes a feature representing the local texture information at the corresponding position.

[0094] In a possible embodiment, extracting feature data corresponding to each mesh vertex from the medical image data (step S430) may include:

[0095] S431: Based on a preset feature extraction operator or deep learning network, extract texture information within a preset range of each mesh vertex from the medical image data as feature data corresponding to each mesh vertex.

[0096] Specifically, based on the position coordinates of each mesh vertex, texture information within a preset range of the position of each mesh vertex in the medical image data can be extracted as corresponding feature data. The preset range can be set as needed, for example, to a spherical region with the position coordinates as the center and a radius of R (where R can be set to the thickness of the heart valve membrane or can be set as needed). This is not limited in this embodiment of the present invention.

[0097] In a possible embodiment, the step S431 may include:

[0098] performing a morphological operation on the first mask image using a morphological algorithm to obtain a second mask image;

[0099] applying the second mask image to the medical image data to obtain second image data of the heart valve region;

[0100] determining a target image region in the medical image data based on the second image data;

[0101] Based on a preset feature extraction operator or a deep learning network, texture information in the projection direction corresponding to each of the mesh vertices is extracted from the target image area as feature data corresponding to each of the mesh vertices.

[0102] Optionally, the first mask image can be dilated or eroded to a certain thickness based on a morphological algorithm to obtain a morphologically transformed second mask image. The second mask image is also a binary image. Applying the second mask image to the medical image data can generate second image data. The image region between the boundary of the visceral valve region of the first image data center and the boundary of the visceral valve region of the second image data center can be used as the target image region. For example, when the second image data is image data obtained by dilation, the target region can be the image region outside the boundary of the visceral valve region of the first image data center and within the boundary of the visceral valve region of the second image data center. When the second image data is image data obtained by erosion, the target region can be the image region inside the boundary of the visceral valve region of the first image data center and outside the boundary of the visceral valve region of the second image data center.

[0103] Optionally, the first mask image can be expanded to a certain thickness based on a morphological algorithm to obtain a second mask image, and then the first mask image can be eroded to a certain thickness based on a morphological algorithm to obtain another second mask image. These two second mask images can be applied to the medical image data respectively to obtain two second image data, and the image area between the boundaries of the visceral valve areas of the two second image data can be used as the target image area.

[0104] It should be noted that in the two optional embodiments described above, the thickness of the morphologically-based expansion or erosion can be set based on actual needs. For example, it can be set to the membrane thickness of the heart valve, or different expansion / erosion thicknesses can be set for different locations. This is not limited in the present embodiment. For example, for the aortic valve region, the thickness can be set to 1 mm, or a larger thickness can be set for the aortic valve root and a smaller thickness can be set for other locations, and so on.

[0105] Specifically, after determining the target image region, texture information in the projection direction corresponding to each mesh vertex can be extracted from the target image region based on a preset feature extraction operator or deep learning network as feature data corresponding to each mesh vertex. The projection direction corresponding to each mesh vertex can be determined based on the topological structure of the surface mesh.

[0106] In another possible embodiment, texture information within a preset distance range in the projection direction corresponding to each mesh vertex can be directly extracted from the medical image data based on a preset feature extraction operator or deep learning network, as feature data corresponding to each mesh vertex. The projection direction corresponding to each mesh vertex can be determined based on the topological structure of the surface mesh, and the preset distance range can be set as needed, which is not limited in this embodiment of the present invention.

[0107] In practical applications, feature data can be extracted from the target image area / medical image data through a variety of feature extraction methods, including but not limited to traditional feature extraction operators and deep learning networks. The traditional feature extraction operators include but are not limited to texture feature extraction operators such as Gray-Level Co-occurrence Matrix (GLCM), Gray-Level Difference Statistics (GLDS), and Gaussian-Markov Random Field (GMRF), and the deep learning network includes but is not limited to convolutional neural networks, such as multi-scale convolutional neural networks. Convolutional neural network is a type of feedforward neural network that includes convolution calculations and has a deep structure. It is one of the representative algorithms of deep learning. Multi-scale convolutional neural network realizes feature extraction of multiple receptive fields by adopting convolution kernels of different sizes and selecting local target images of different sizes on the basis of convolutional neural network, and uses multi-scale image features to provide image structure information at different levels.

[0108] It should be noted that the method of extracting feature data using traditional feature extraction operators and deep learning networks can refer to the existing technology, and the embodiments of the present invention will not be described in detail here.

[0109] It can be understood that the embodiment of the present invention projects structural information onto the surface grid through a grid projection segmentation strategy, which can reduce the error caused by the structure thickness and at the same time reduce the algorithm complexity of the image segmentation, thereby improving the accuracy of image segmentation and the accuracy of numerical simulation.

[0110] S440: Constructing a graph model based on the surface mesh and feature data corresponding to each mesh vertex.

[0111] In an embodiment of the present invention, a graph cut algorithm can be used to define each mesh vertex of the surface mesh as a vertex of a graph, and a graph model can be generated to transform the image segmentation problem into an energy minimization problem. The graph model consists of a vertex set V and an edge set E, and includes two types of vertices and two types of edges: one type of vertex is a normal vertex, which is a mesh vertex of the surface mesh. Every two adjacent normal vertices are connected by an edge, called neighborhood links (n-links); the other type is two terminal vertices independent of normal vertices, representing the foreground S and background T of the segmentation target. Every terminal vertex is connected to each normal vertex by an edge, called terminal links (t-links).

[0112] In this embodiment of the present invention, the weight of each edge in the graph model can be determined based on the feature data corresponding to each mesh vertex extracted in step S430. Specifically, the weight of each edge comes from two parts: a region term and a boundary term. The region term encodes the grayscale distribution of different categories and connects ordinary vertices and terminal vertices (i.e., the segmented foreground and background); the boundary term maintains the continuity between adjacent ordinary vertices and connects the edges of adjacent ordinary vertices.

[0113] Specifically, a deep learning network can be used to fit the mapping relationship between the feature data of the mesh vertices and the n-link weights and t-link weights to estimate the n-link and t-link weights. The deep learning network may include but is not limited to a convolutional neural network, such as a multi-scale convolutional neural network.

[0114] In practical applications, a deep learning network model can be pre-trained to extract feature data corresponding to each mesh vertex from medical image data and estimate the weights of n-links and t-links based on the feature data corresponding to each mesh vertex. It should be noted that the training method for the deep learning network model is prior art and will not be further described in detail in the present embodiment.

[0115] S450: Solve the graph model using a graph cut algorithm to obtain a multi-classification result of the image grid corresponding to the heart valve region.

[0116] In an embodiment of the present invention, a calculation formula for segmentation energy can be derived by integrating the region term and the boundary term based on methods including but not limited to a maximum flow / minimum cut algorithm. An optimal solution that minimizes energy can be obtained through multiple iterative calculations. The surface mesh can then be classified based on the optimal solution of the graphical model to obtain a multi-classification result of the image mesh corresponding to the heart valve region, serving as the image segmentation result corresponding to the heart valve region. The multi-classification result of the image mesh can be a surface mesh with classification information, and the classification information can be used to distinguish different heart regions and their corresponding surface mesh structures.

[0117] It should be noted that the specific content of solving the optimal solution of the graphical model through the maximum flow / minimum cut algorithm can be referred to the existing technology, and the embodiments of the present invention will not be repeated here.

[0118] It can be understood that the embodiments of the present invention combine traditional algorithms with deep learning algorithms, integrating the contour representation of graph nodes with a multi-scale image patch strategy, and using the proposed multi-scale convolutional neural network to learn edge weights, thereby transforming the image segmentation problem into a minimum cut problem for a graph model. By combining traditional and deep learning algorithms, they can complement each other, effectively addressing the problems of traditional algorithms' single features and poor robustness, as well as the difficulty of deep learning algorithms in achieving accurate boundary segmentation for small objects, thereby improving the accuracy and robustness of image segmentation.

[0119] S230: Generate a dynamic model of the heart valve region according to the multiple image segmentation results.

[0120] In an embodiment of the present invention, each image segmentation result includes surface mesh structures corresponding to multiple cardiac regions located in the heart valve region, where the multiple cardiac regions may include at least one valve and corresponding related regions. For example, each image segmentation result corresponding to the aortic valve region may include surface mesh structures corresponding to three aortic valves (one or two in a pathological state) and a surface mesh structure corresponding to the aorta.

[0121] In practical applications, the image segmentation result can be a surface grid structure with classification information corresponding to the heart valve area (i.e., an image grid multi-classification result). The classification information can be used to distinguish different heart parts (including at least one valve and corresponding related parts). The surface grid structure corresponding to each heart part can be determined based on the classification information.

[0122] In an embodiment of the present invention, generating a dynamic model of the heart valve region according to the multiple image segmentation results (step S230) may include:

[0123] for each heart region, determining a plurality of surface mesh structures corresponding to the heart region, and generating a dynamic model of the heart region according to the plurality of surface mesh structures;

[0124] A dynamic model of the heart valve region is generated based on the dynamic models of the respective heart regions.

[0125] Optionally, generating the dynamic model of the heart region may include:

[0126] Optimizing and reconstructing each of the surface grid structures to form a corresponding surface structure;

[0127] Respectively performing volume reconstruction and spatial meshing on the area surrounded by each surface structure to form a corresponding volume structure;

[0128] A dynamic model of the heart region is generated based on the volume structures corresponding to the surface mesh structures.

[0129] In the embodiment of the present invention, since each image segmentation result includes a surface grid structure corresponding to each heart part, multiple surface grid structures corresponding to the heart part can be obtained according to the multiple image segmentation results.

[0130] In practical applications, the surface mesh structures in each image segmentation result can be first refined to achieve an accuracy that can support accurate simulation calculations, and ensure that the number of mesh vertices in each refined surface mesh structure is the same and can correspond one to one. The refined surface mesh structure is then used to model each part of the heart separately to obtain the corresponding dynamic model.

[0131] Specifically, because each surface mesh structure can be used to reconstruct a volumetric structure of the heart region, multiple models of the heart region at different stages within a cardiac cycle can be obtained. These multiple models can then be used to generate a dynamic model of the heart region that continuously changes during the cardiac cycle. Furthermore, the dynamic models of the various heart regions can be combined to generate a dynamic model of the heart valve region.

[0132] Exemplarily, for the aortic valve region (i.e., the valved aortic root region), multiple corresponding surface grid structures can be determined for each independent aortic valve and aorta, and each surface grid structure can be reconstructed with a certain thickness to form a corresponding surface structure. The area surrounded by the surface structure is then volume reconstructed and spatially gridded to form a corresponding volume structure, and then a corresponding dynamic model is generated. The dynamic models corresponding to the independent aortic valves and aorta are then combined to obtain a dynamic model of the aortic valve region (i.e., the valved aortic root region).

[0133] It should be noted that the thickness of the reconstructed surface mesh structures can be set according to actual needs. For example, it can be set to the membrane thickness of the heart valve, or different reconstruction thicknesses can be set for different locations, etc. This embodiment of the present invention is not limited to this. For example, for the aortic valve, the thickness can be set to 1 mm, or a larger thickness can be set for the aortic valve root and a smaller thickness can be set for other locations, etc.

[0134] It should be noted that both the surface structure reconstruction method and the volume structure reconstruction method are prior arts and will not be described in detail in the embodiments of the present invention.

[0135] It should be noted that, in some possible embodiments, the surface grid structure may not be reconstructed to form a volume structure, but the dynamic model of the heart region may be directly generated using the multiple surface grid structures.

[0136] It can be understood that the embodiment of the present invention independently models each heart part (especially each valve) in the heart valve region, so that each valve in the final generated dynamic model can move independently, thereby obtaining a more accurate dynamic model.

[0137] S240: Based on the dynamic model, determine the mechanical distribution information of the heart valve region using a non-steady-state simulation method.

[0138] In an embodiment of the present invention, the heart valve area can be defined as a solid domain, and the blood flow can be defined as a fluid domain. Numerical simulation calculations are performed on the solid domain and the fluid domain within one cardiac cycle respectively to obtain the structural mechanical parameters and hemodynamic parameters of each position in the heart valve area.

[0139] In a possible embodiment, determining the mechanical distribution information of the heart valve region using a non-steady-state simulation method based on the dynamic model may include:

[0140] Performing solid domain numerical simulation based on the dynamic model to obtain structural mechanical parameters of various positions in the heart valve region; and / or,

[0141] obtaining a blood flow velocity in the heart valve region as a boundary condition;

[0142] Fluid domain numerical simulation is performed based on the dynamic model and the blood flow velocity to obtain hemodynamic parameters of the heart valve region.

[0143] Specifically, for the solid domain, the dynamic model can be used as input to perform numerical simulation calculations using numerical simulation methods including but not limited to Newton's method, or using commercial simulation software such as Abaqus, to obtain structural mechanical parameters of various locations in the heart valve region at different times within a cardiac cycle. These structural mechanical parameters may include stress, shear force, transvalvular pressure difference, etc., ultimately forming a cyclical mechanical image that changes with time. The initial state at each moment is initialized with the final state at the previous moment.

[0144] It can be understood that since the heart is always beating, the numerical simulation based on a dynamic model of a complete cardiac cycle proposed in the embodiment of the present invention is more in line with the actual situation than the steady-state simulation based on a single model. It can also provide important information that cannot be obtained by steady-state simulation, including but not limited to stress information, and provides reliable data support for the subsequent determination of abnormal locations.

[0145] Specifically, for the fluid domain, a single-frame or multi-frame model in the dynamic model can be pre-selected, and the model of each valve can be used as the morphological boundary condition, and the blood flow velocity in the heart valve area can be used as the hemodynamic boundary condition. Corresponding boundary conditions can be set for different models according to the cardiac cycle corresponding to the model, that is, when the valve is open during systole, it is set to have blood flow input and the corresponding blood flow velocity, and when the valve is closed during diastole, it is set to have no blood flow input. The selected single-frame or multi-frame model is used as input, and fluid dynamics simulation calculations are performed using commercial simulation software such as Ansys to form a periodic hemodynamic image that changes with time. Based on the hemodynamic image, hemodynamic parameters such as pressure and shear force in the heart valve area can be determined. The blood flow velocity in the heart valve area is a periodically changing blood flow velocity, and the initial state of each moment in the periodic hemodynamic image is initialized with the final state of the previous moment.

[0146] Specifically, the blood flow velocity in the heart valve region can be measured or calculated using various methods in the prior art, or the standard blood flow velocity in the heart valve region can be used as the blood flow velocity in the heart valve region, and the embodiments of the present invention are not limited thereto. For example, the blood flow velocity in the heart valve region can be measured using angiography, for example, by measuring the flow velocity of a contrast agent in the heart valve region as the blood flow velocity in the heart valve region, or by calculating the average blood flow velocity during the contrast agent filling process based on angiography image sequences of the heart valve region as the blood flow velocity in the heart valve region, etc.

[0147] It can be understood that since each valve in the dynamic model can move independently, the mechanical distribution information obtained through numerical simulation analysis is more consistent with the actual situation and has higher accuracy.

[0148] It should be noted that, in the absence of multiple medical image data within a complete cardiac cycle, numerical simulation of the fluid domain at a single moment can be achieved to obtain the hemodynamic parameters of the heart valve region. Figure 2 The method shown constructs a single model structure of the heart valve area based on single medical image data, and uses dynamic blood flow input to perform full-cycle numerical simulation calculations on the single model structure to obtain the hemodynamic parameters of the heart valve area, but cannot obtain the structural mechanics parameters of the solid domain.

[0149] In one possible embodiment, the Figure 7 , the method may further comprise the following steps:

[0150] S250: determining a location with abnormal mechanical distribution in the heart valve region according to the mechanical distribution information;

[0151] S260: Prompt the location where the mechanical distribution is abnormal.

[0152] Specifically, it is possible to determine whether there is a site with abnormal mechanical distribution in the heart valve region based on the structural mechanical parameters and / or hemodynamic parameters of each position in the heart valve region. If so, a prompt will be given for the site with abnormal mechanical distribution.

[0153] Specifically, it is possible to determine whether each mechanical parameter is abnormal based on the preset mechanical index data, and determine the site with abnormal mechanical parameters as an abnormal site. For example, it is possible to determine whether the stress at each location in the heart valve region is greater than the preset stress index, whether the shear force is within the preset index range, and whether the transvalvular pressure difference is within the preset pressure difference index range, etc., and determine the site where the stress is greater than the preset stress index, the shear force is outside the preset index range, or the transvalvular pressure difference is outside the preset pressure difference index range as an abnormal site, and provide a prompt. The mechanical index data can be set according to actual conditions, for example, it can be set according to the standard data corresponding to the heart valve region in a normal heart, and the embodiment of the present invention does not impose any restrictions on this.

[0154] It can be understood that based on the mechanical analysis results of the heart valve area, the mechanical changes of the heart valve area during a cardiac cycle can be determined, and then it can be judged whether there are abnormal structures in the heart valve area, which can further indicate the material lesions of the valve tissue and ultimately locate the abnormal site, providing a basis for the selection of valve prostheses in TAVR surgery, realizing patient-specific preoperative planning plans, and then guiding the surgical process.

[0155] In summary, according to the mechanical analysis method of the heart valve area according to an embodiment of the present invention, multiple medical image data within a cardiac cycle are segmented to obtain multiple image segmentation results corresponding to the heart valve area, and a dynamic model of the heart valve area is generated based on the multiple image segmentation results. By performing non-steady-state numerical simulation based on the dynamic model, mechanical distribution information that is more in line with the actual situation can be obtained, which can truly reflect the mechanical changes of the heart valve area during movement, thereby improving the accuracy of the mechanical analysis results.

[0156] Reference Manual Figure 8 , which shows the structure of a mechanical analysis device 800 for a heart valve region provided by one embodiment of the present invention. Figure 8 As shown, the apparatus 800 may include:

[0157] The medical image data acquisition module 810 is configured to acquire a plurality of medical image data including a heart valve region corresponding to different moments in a cardiac cycle;

[0158] An image segmentation module 820 is configured to perform image segmentation processing on each of the medical image data to obtain a plurality of image segmentation results corresponding to the heart valve region;

[0159] A dynamic model generation module 830 is configured to generate a dynamic model of the heart valve region based on the multiple image segmentation results;

[0160] The mechanical distribution information determination module 840 is configured to determine the mechanical distribution information of the heart valve region based on the dynamic model using a non-steady-state simulation method.

[0161] In a possible embodiment, the apparatus 800 may further include:

[0162] an abnormal part determination module, configured to determine a part of the heart valve region with abnormal mechanical distribution according to the mechanical distribution information;

[0163] The prompt module is used to prompt the parts with abnormal mechanical distribution.

[0164] It should be noted that the devices provided in the above embodiments are only illustrated by the division of the above functional modules when implementing their functions. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the devices provided in the above embodiments and the corresponding method embodiments are based on the same concept. The specific implementation process is detailed in the corresponding method embodiments and will not be repeated here.

[0165] One embodiment of the present invention also provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the mechanical analysis method of the heart valve area provided in the above-mentioned method embodiment.

[0166] Specifically, the memory can be used to store software programs and modules. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory. The memory can mainly include a program storage area and a data storage area. The program storage area can store an operating system, application programs required for functions, etc.; the data storage area can store data created based on the use of the device, etc. In addition, the memory can include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory can also include a memory controller to provide the processor with access to the memory.

[0167] In a specific embodiment, Figure 9 The hardware structure diagram of an electronic device for implementing the mechanical analysis method of the heart valve region provided by an embodiment of the present invention is shown. The electronic device can be a computer terminal, a mobile terminal or other device. The electronic device can also participate in or include the mechanical analysis device of the heart valve region provided by an embodiment of the present invention. Figure 9 As shown, the electronic device 900 may include one or more computer-readable storage media memories 910, one or more processing core processors 920, an input unit 930, a display unit 940, a radio frequency (RF) circuit 950, a wireless fidelity (WiFi) module 960, and a power supply 970. Those skilled in the art will understand that Figure 9 The electronic device structure shown in the figure does not constitute a limitation on the electronic device 900, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0168] The memory 910 can be used to store software programs and modules. The processor 920 executes various functional applications and data processing by running or executing the software programs and modules stored in the memory 910 and calling the data stored in the memory 910. The memory 910 may mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created based on the use of the electronic device, etc. In addition, the memory 910 may include a high-speed random access memory and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory 910 may also include a memory controller to provide the processor 920 with access to the memory 910.

[0169] The processor 920 is the control center of the electronic device 900. It connects the various parts of the entire electronic device using various interfaces and lines. By running or executing software programs and / or modules stored in the memory 910 and accessing data stored in the memory 910, it performs various functions of the electronic device 900 and processes data, thereby monitoring the entire electronic device 900. The processor 920 can be a central processing unit, or other general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0170] The input unit 930 may be configured to receive input digital or character information and generate keyboard, mouse, joystick, optical, or trackball signal input related to user settings and function control. Specifically, the input unit 930 may include a touch-sensitive surface 931 and other input devices 932. Specifically, the touch-sensitive surface 931 may include, but is not limited to, a touchpad or a touch screen, and the other input devices 932 may include, but are not limited to, one or more of a physical keyboard, function keys (such as a volume control button, an on / off button, etc.), a trackball, a mouse, a joystick, and the like.

[0171] The display unit 940 can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces of the electronic device. These graphical user interfaces can be composed of graphics, text, icons, videos, or any combination thereof. The display unit 940 may include a display panel 941. Optionally, the display panel 941 can be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

[0172] The RF circuit 950 can be used to receive and send signals during information transmission or calls. In particular, after receiving downlink information from the base station, it is handed over to one or more processors 920 for processing; in addition, uplink data is sent to the base station. Generally, the RF circuit 950 includes but is not limited to an antenna, at least one amplifier, a tuner, one or more oscillators, a subscriber identity module (SIM) card, a transceiver, a coupler, a low noise amplifier (LNA), a duplexer, etc. In addition, the RF circuit 950 can also communicate with the network and other devices through wireless communication. The wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobile Communication (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), email, Short Messaging Service (SMS), etc.

[0173] WiFi is a short-range wireless transmission technology. The electronic device 900 can help users send and receive emails, browse web pages, and access streaming media through the WiFi module 960. It provides users with wireless broadband Internet access. Figure 9 A WiFi module 960 is shown, but it is understandable that it is not an essential component of the electronic device 900 and can be omitted as needed without changing the essence of the invention.

[0174] The electronic device 900 also includes a power supply 970 (e.g., a battery) for supplying power to various components. Preferably, the power supply can be logically connected to the processor 920 via a power management system, thereby enabling the power management system to manage charging, discharging, and power consumption. The power supply 970 can also include any of one or more DC or AC power supplies, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and other components.

[0175] It should be noted that, although not shown, the electronic device 900 may further include a Bluetooth module, etc., which will not be described in detail here.

[0176] One embodiment of the present invention also provides a computer-readable storage medium, which can be set in an electronic device to store at least one instruction or at least one program related to implementing a mechanical analysis method for a heart valve area. The at least one instruction or the at least one program is loaded and executed by the processor to implement the mechanical analysis method for the heart valve area provided in the above-mentioned method embodiment.

[0177] Optionally, in an embodiment of the present invention, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store program codes.

[0178] One embodiment of the present invention also provides a computer program product, which includes a computer program / instructions. When the computer program product is run on an electronic device, the computer program / instructions are loaded and executed by a processor to implement the steps of the mechanical analysis method of the heart valve area provided in the various optional embodiments above.

[0179] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not represent the superiority or inferiority of the embodiments. The above description is of specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0180] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences from other embodiments. In particular, the device embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

[0181] Those skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware, or by a program to instruct the relevant hardware, and the program may be stored in a computer-readable storage medium, which may be a read-only memory, a disk, or an optical disk, etc.

[0182] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A mechanical analysis method for a heart valve region, characterized in that: include: S210: Acquire a plurality of medical image data including a heart valve region corresponding to different moments in a cardiac cycle; S220: Performing image segmentation processing on each of the medical image data using a graph cut algorithm to obtain multiple image segmentation results corresponding to the heart valve region; the image segmentation results include surface mesh structures corresponding to multiple heart parts, the surface mesh structures having classification information for distinguishing different heart parts, the multiple heart parts including at least one valve and corresponding related parts; S230: generating a dynamic model of the heart valve region according to the multiple image segmentation results; S240: Determine mechanical distribution information of the heart valve region using a non-steady-state simulation method based on the dynamic model; The step S230 includes: for each heart region, determining a plurality of surface mesh structures corresponding to the heart region, and generating a dynamic model of the heart region according to the plurality of surface mesh structures; generating a dynamic model of the heart valve region based on the dynamic models of each of the heart parts; The step S220 includes: S221: extracting first image data of the heart valve region from each medical image data respectively; S222: performing image segmentation processing on the first image data of the heart valve region to obtain an image segmentation result corresponding to the heart valve region; Wherein, the step S222 includes: S410: performing image segmentation processing on the first image data to obtain a first mask image of the heart valve region; S420: generating a surface mesh of the first mask image using a marching cubes algorithm, where the surface mesh includes a plurality of mesh vertices; S430: Extracting feature data corresponding to each mesh vertex from the medical image data; S440: Constructing a graph model based on the surface mesh and feature data corresponding to each mesh vertex; S450: Solving the graph model using a graph cut algorithm to obtain a multi-classification result of the image grid corresponding to the heart valve region.

2. The method according to claim 1, characterized in that The step S430 includes: S431: Based on a preset feature extraction operator or deep learning network, extract texture information within a preset range of each mesh vertex from the medical image data as feature data corresponding to each mesh vertex.

3. The method according to claim 2, characterized in that The S431 step includes: performing a morphological operation on the first mask image using a morphological algorithm to obtain a second mask image; applying the second mask image to the medical image data to obtain second image data of the heart valve region; determining a target image region in the medical image data based on the second image data; Based on a preset feature extraction operator or a deep learning network, texture information in the projection direction corresponding to each of the mesh vertices is extracted from the target image area as feature data corresponding to each of the mesh vertices.

4. The method according to claim 1, wherein The generating of the dynamic model of the heart region comprises: Optimizing and reconstructing each of the surface grid structures to form a corresponding surface structure; Respectively performing volume reconstruction and spatial meshing on the area surrounded by each surface structure to form a corresponding volume structure; A dynamic model of the heart region is generated based on the volume structures corresponding to the surface mesh structures.

5. The method according to claim 1, wherein The step S240 includes: Performing solid domain numerical simulation based on the dynamic model to obtain structural mechanical parameters of various positions in the heart valve region; and / or, obtaining a blood flow velocity in the heart valve region as a boundary condition; Fluid domain numerical simulation is performed based on the dynamic model and the blood flow velocity to obtain hemodynamic parameters of the heart valve region.

6. The method according to claim 1, characterized in that The method further comprises: S250: determining a location with abnormal mechanical distribution in the heart valve region according to the mechanical distribution information; S260: Prompt the location where the mechanical distribution is abnormal.

7. A mechanical analysis device for a heart valve region, characterized in that: include: A medical image data acquisition module, configured to acquire a plurality of medical image data including a heart valve region corresponding to different moments in a cardiac cycle; an image segmentation module, configured to perform image segmentation processing on each of the medical image data using a graph cut algorithm to obtain a plurality of image segmentation results corresponding to the heart valve region; the image segmentation results comprising surface mesh structures corresponding to a plurality of heart regions, the surface mesh structures carrying classification information for distinguishing different heart regions, the plurality of heart regions comprising at least one valve and corresponding related regions; a dynamic model generating module, configured to generate a dynamic model of the heart valve region according to the plurality of image segmentation results; a mechanical distribution information determination module, configured to determine the mechanical distribution information of the heart valve region based on the dynamic model and using a non-steady-state simulation method; The step of generating a dynamic model of the heart valve region according to the multiple image segmentation results includes: for each heart region, determining a plurality of surface mesh structures corresponding to the heart region, and generating a dynamic model of the heart region according to the plurality of surface mesh structures; generating a dynamic model of the heart valve region based on the dynamic models of each of the heart parts; The step of performing image segmentation processing on each of the medical image data using a graph cut algorithm to obtain multiple image segmentation results corresponding to the heart valve region includes: For each piece of medical image data, extract first image data of the heart valve region from the medical image data; performing image segmentation processing on the first image data of the heart valve region to obtain an image segmentation result corresponding to the heart valve region; The performing image segmentation processing on the first image data of the heart valve region to obtain an image segmentation result corresponding to the heart valve region includes: performing image segmentation processing on the first image data to obtain a first mask image of the heart valve region; generating a surface mesh of the first mask image using a marching cubes algorithm, the surface mesh comprising a plurality of mesh vertices; extracting feature data corresponding to each of the mesh vertices from the medical image data; Constructing a graph model based on the surface mesh and feature data corresponding to each of the mesh vertices; The graph model is solved using a graph cut algorithm to obtain a multi-classification result of the image grid corresponding to the heart valve region.

8. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the mechanical analysis method of the heart valve area as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the mechanical analysis method of the heart valve region according to any one of claims 1 to 6.

Images