Sparse representation method, device, equipment, medium and program product of blood vessel model

Abstract

The application relates to the medical and engineering cross technical field, in particular to a blood vessel model sparse expression method, device, equipment, medium and program product, wherein the method comprises the following steps: constructing a three-dimensional blood vessel model of a target blood vessel; extracting a blood vessel tree and an extraluminal cross section of the target blood vessel based on the three-dimensional blood vessel model; fitting the blood vessel tree based on the extraluminal cross section to obtain a blood vessel surface fitting result; and establishing a tree structure of the blood vessel tree by using the blood vessel surface fitting result, so as to obtain sparse expression of the three-dimensional blood vessel model based on the tree structure. Thus, the three-dimensional model of the blood vessel is expressed by a polygonal mesh in the related art, and it is difficult to perform real-time physical simulation in a dense collision detection scene, the real-time performance is poor, and the actual use requirements cannot be met.

Description

Technical Field

[0001] This application relates to the field of medical-engineering interdisciplinary technology, and in particular to a sparse representation method, device, equipment, medium and program product for a vascular model. Background Technology

[0002] Angiography is an interventional medical examination and treatment method used to assess and treat diseases and abnormalities in the vascular system. Major angiography techniques include: DSA (Digital Subtraction Angiography), MRA (Magnetic Resonance Angiography), CTA (Coronary Computed Tomography), and transcatheter angiography. Angiography can assess vascular diseases such as arterial stenosis, aneurysms, varicose veins, and thrombosis, providing physicians with detailed information on vascular anatomy and lesions, and helping to guide appropriate treatment and interventional procedures.

[0003] Vascular geometry modeling refers to the process of geometrically modeling the anatomical structure of blood vessels. There are two main methods for constructing the tree structure of a vascular model: one is to start from a few seed points and grow vascular regions or skeletons; the other is to use perceptual grouping technology to construct a tree structure from detected vascular segments. The tree structure of a vascular model includes the geometric attributes of vascular branches, such as length, diameter, bifurcation angle, and spatial orientation. Commonly used data structures include symbolic trees, binary encoding, and implicit surfaces.

[0004] However, in related technologies, the three-dimensional model of blood vessels is mostly expressed using polygonal meshes, which makes it difficult to perform real-time physical simulation in dense collision detection scenarios. This results in poor real-time performance, which cannot meet the actual usage requirements and urgently needs improvement. Summary of the Invention

[0005] This application provides a sparse representation method, apparatus, device, medium, and program product for blood vessel models to solve the problems in related technologies, such as the fact that three-dimensional blood vessel models are mostly represented by polygonal meshes, which makes it difficult to perform real-time physical simulation in dense collision detection scenarios, resulting in poor real-time performance and failure to meet practical application needs.

[0006] The first aspect of this application provides a sparse representation method for a blood vessel model, comprising the following steps: constructing a three-dimensional blood vessel model of a target blood vessel; extracting the blood vessel tree and the outer cross section of the lumen of the target blood vessel based on the three-dimensional blood vessel model; fitting the blood vessel tree with cylinders and spheres based on the outer cross section of the lumen to obtain a blood vessel surface fitting result; establishing a tree structure of the blood vessel tree using the blood vessel surface fitting result, so as to obtain a sparse representation of the three-dimensional blood vessel model based on the tree structure.

[0007] Optionally, in one embodiment of this application, constructing a three-dimensional vascular model of the target blood vessel includes: acquiring an initial angiography image of the target blood vessel; segmenting the initial angiography image to obtain an angiography image of the target blood vessel; extracting the vascular structure of the angiography image of the target blood vessel; and reconstructing the three-dimensional vascular model of the target blood vessel based on the vascular results.

[0008] Optionally, in one embodiment of this application, the step of extracting the vascular tree and external cross-section of the target blood vessel based on the three-dimensional vascular model includes: extracting at least one initial path of the target blood vessel based on the three-dimensional vascular model; obtaining at least one vascular centerline of the target blood vessel based on the at least one initial path; generating the vascular tree using the at least one vascular centerline, and calculating the external cross-section of the lumen.

[0009] Optionally, in one embodiment of this application, the step of fitting the vascular tree with cylinders and spheres based on the outer cross section of the lumen to obtain the vascular surface fitting result includes: discretizing the vascular tree to obtain at least one cylinder in the vascular tree; connecting the at least one cylinder with at least one sphere based on the outer cross section of the lumen to obtain the cylinder expression and sphere expression of the vascular tree, and generating the vascular surface fitting result.

[0010] Optionally, in one embodiment of this application, the step of establishing a tree structure of the blood vessel tree using the blood vessel surface fitting results to obtain a sparse representation of the three-dimensional blood vessel model based on the tree structure includes: establishing a tree structure of the blood vessel tree using the blood vessel surface fitting results based on the nodes and edges of the three-dimensional blood vessel model; and generating blood vessel information for the sparse representation of the three-dimensional blood vessel model using the cylindrical and spherical representations of the tree structure.

[0011] A second aspect of this application provides a sparse representation device for a blood vessel model, comprising: a construction module for constructing a three-dimensional blood vessel model of a target blood vessel; an extraction module for extracting a blood vessel tree and an outer cross section of the target blood vessel based on the three-dimensional blood vessel model; and a generation module for fitting the blood vessel tree to cylinders and spheres based on the outer cross section of the lumen to obtain a blood vessel surface fitting result, and establishing a tree structure of the blood vessel tree using the blood vessel surface fitting result to obtain a sparse representation of the three-dimensional blood vessel model based on the tree structure.

[0012] Optionally, in one embodiment of this application, the construction module includes: an acquisition unit for acquiring an initial angiography image of the target blood vessel; a first generation unit for segmenting the initial angiography image to obtain a target blood vessel angiography image; and a construction unit for extracting the vascular structure of the target blood vessel angiography image and reconstructing the three-dimensional vascular model of the target blood vessel based on the vascular results.

[0013] Optionally, in one embodiment of this application, the extraction module includes: an extraction unit for extracting at least one initial path of the target blood vessel based on the three-dimensional blood vessel model; a second generation unit for obtaining at least one blood vessel centerline of the target blood vessel based on the at least one initial path; and a third generation unit for generating the blood vessel tree using the at least one blood vessel centerline and calculating the external cross-section of the lumen.

[0014] Optionally, in one embodiment of this application, the generation module includes: a discretization unit for discretizing the vascular tree to obtain at least one cylinder in the vascular tree; and a fourth generation unit for connecting the at least one cylinder with at least one sphere based on the outer cross section of the lumen to obtain the cylinder representation and sphere representation of the vascular tree, and generating the vascular surface fitting result.

[0015] Optionally, in one embodiment of this application, the generation module includes: a building unit, used to build a tree structure of the blood vessel tree based on the nodes and edges of the three-dimensional blood vessel model and using the blood vessel surface fitting results; and a fifth generation unit, used to generate blood vessel information for sparse representation of the three-dimensional blood vessel model using the cylindrical and spherical expressions of the tree structure.

[0016] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the sparse representation method of the vascular model as described in the above embodiments.

[0017] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the sparse representation method of the blood vessel model described above.

[0018] A fifth aspect of this application provides a computer program product, including a computer program that, when executed, implements the sparse representation method of the vascular model described above.

[0019] This application embodiment can extract the vascular tree and lumen cross-section from the constructed 3D vascular model of the target blood vessel, and achieve cylindrical and spherical fitting of the vascular tree based on the lumen cross-section. Then, the fitting results are used to establish a tree-like structure of the vascular tree, completing the sparse representation of the 3D vascular model. This reduces the computational load and complexity in the virtual vascular surgery simulation process, and improves the efficiency of physical simulation and computational calculation of blood vessels. Therefore, it solves the problems in related technologies where 3D vascular models are mostly expressed using polygonal meshes, making real-time physical simulation difficult in dense collision detection scenarios, resulting in poor real-time performance and failing to meet practical application requirements.

[0020] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0021] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0022] Figure 1 This is a flowchart of a sparse representation method for a blood vessel model according to an embodiment of this application;

[0023] Figure 2 A schematic diagram of the effect of a three-dimensional blood vessel model reconstructed from CTA images using the level set algorithm according to an embodiment of this application;

[0024] Figure 3 This is a flowchart illustrating the extraction of the vascular tree and external cross-section of a target blood vessel using a bidirectional minimum path propagation method according to an embodiment of this application.

[0025] Figure 4 A block diagram illustrating the effect of the extracted target blood vessel tree and lumen cross-section according to an embodiment of this application;

[0026] Figure 5 This is a flowchart illustrating the fitting of cylinders and spheres in a vascular tree using the RANSAC (Random Sample Consensus) algorithm according to an embodiment of this application.

[0027] Figure 6 This is a block diagram illustrating the fitting effect of the RANSAC algorithm on the blood vessel surface based on cylinders and spheres according to an embodiment of this application;

[0028] Figure 7 This is a block diagram illustrating the tree structure of a blood vessel tree according to an embodiment of this application;

[0029] Figure 8 A flowchart illustrating the working principle of a sparse representation method for a blood vessel model according to an embodiment of this application;

[0030] Figure 9 This is a block diagram of a sparse expression device for a blood vessel model according to an embodiment of this application;

[0031] Figure 10 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0032] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0033] The following description, with reference to the accompanying drawings, illustrates the sparse representation, apparatus, device, medium, and program product of a blood vessel model according to embodiments of this application. Addressing the problem mentioned in the background art that 3D blood vessel models are often represented using polygonal meshes, making real-time physical simulation difficult in dense collision detection scenarios, resulting in poor real-time performance and failing to meet practical application requirements, this application provides a sparse representation method for blood vessel models. In this method, the vascular tree and external cross-section of the lumen can be extracted from a constructed 3D blood vessel model. Based on the external cross-section, cylindrical and spherical fitting of the vascular tree is performed, and the fitting results are used to establish a tree-like structure of the vascular tree, completing the sparse representation of the 3D blood vessel model. This reduces the computational load and complexity in virtual vascular surgery simulation, improving the efficiency of physical simulation and computational calculations for blood vessels. Therefore, this solves the problems in related technologies where 3D blood vessel models are often represented using polygonal meshes, making real-time physical simulation difficult in dense collision detection scenarios, resulting in poor real-time performance and failing to meet practical application requirements.

[0034] Specifically, Figure 1 This is a flowchart of a sparse representation method for a blood vessel model provided according to an embodiment of this application.

[0035] like Figure 1As shown, the sparse representation method of this blood vessel model includes the following steps:

[0036] In step S101, a three-dimensional vascular model of the target blood vessel is constructed.

[0037] As one implementation method, embodiments of this application can construct a three-dimensional vascular model of the target blood vessel. For example, embodiments of this application can construct a three-dimensional vascular model of the target blood vessel based on CTA images.

[0038] Optionally, in one embodiment of this application, constructing a three-dimensional vascular model of the target blood vessel includes: acquiring an initial angiography image of the target blood vessel; segmenting the initial angiography image to obtain an angiography image of the target blood vessel; extracting the vascular structure of the angiography image of the target blood vessel; and reconstructing a three-dimensional vascular model of the target blood vessel based on the vascular results.

[0039] It is understood that angiographic images may include, but are not limited to, digital subtraction angiography, magnetic resonance angiography, CTA, and transcatheter angiography, etc., and this application does not impose specific limitations. For example, embodiments of this application may use CTA images as the initial angiographic images for illustration.

[0040] Furthermore, in the embodiments of this application, the segmentation method of the initial angiography image is mainly used to accurately extract the vascular structure of the target angiography image from the medical image. Common segmentation methods may include, but are not limited to: threshold-based segmentation; edge detection algorithms (such as Sobel, Canny, etc., which are not specifically limited in this application) to detect vascular edges; region-based segmentation; separation of blood vessels from complex backgrounds based on horizontal line transformation (such as geometric analysis methods such as Hough transform, which are not specifically limited in this application); deep learning-based methods, etc., which are not specifically limited in this application.

[0041] In some embodiments, the initial angiography image of the target blood vessel can be segmented to obtain an angiography image of the target blood vessel, and then the blood vessel structure can be extracted to reconstruct a three-dimensional blood vessel model of the target blood vessel.

[0042] For example, this application embodiment takes the level set algorithm as an example. The level set algorithm is a blood vessel segmentation method based on an active contour model, which can use local intensity and smoothness information to extract blood vessel structures.

[0043] Specifically, in this embodiment, the level set function is a real-valued function defined in the image domain, and its zero level set is the blood vessel contour. Furthermore, in this embodiment, the level set function can evolve through a partial differential equation, thereby making the blood vessel contour gradually approximate the actual blood vessel boundary.

[0044] Furthermore, in this embodiment, the level set algorithm requires an initial contour as input, and then adjusts the shape and position of the contour based on internal and external energy terms. The internal energy term ensures the smoothness and continuity of the contour, while the external energy term attracts the contour towards the blood vessel edge based on image features. Furthermore, this embodiment can embed an implicitly expressed initial contour surface {x|φ(x,t)=0} into the image, and then repeatedly deform the surface to include the region of interest. Each implicitly defined point on the surface is deformed along the normal path of the local surface driven by a velocity function. The velocity function F(x,t) is a key factor determining the surface motion in the level set algorithm; it depends on the coordinate x in the image and the simulation time t. In addition, the velocity function integrates the average curvature of the local surface and the intensity information of the local image, enabling the surface to move smoothly and preventing the omission of insignificant regions in the image due to weak connections at the boundary of the region of interest. Figure 2 As shown, the embodiments of this application reconstruct a three-dimensional blood vessel model based on CTA images using the level set algorithm.

[0045] In step S102, the vascular tree and external cross-section of the target blood vessel are extracted based on the three-dimensional vascular model.

[0046] As one possible approach, embodiments of this application can extract the vascular tree and external cross-section of the target blood vessel based on a three-dimensional vascular model.

[0047] Optionally, in one embodiment of this application, extracting the vascular tree and external cross-section of the target blood vessel based on a three-dimensional vascular model includes: extracting at least one initial path of the target blood vessel based on the three-dimensional vascular model; obtaining at least one vascular centerline of the target blood vessel based on the at least one initial path; generating a vascular tree using the at least one vascular centerline; and calculating the external cross-section of the lumen.

[0048] It is understood that, in this embodiment of the application, the vascular centerline can be defined as the weighted shortest path between two endpoints. The extraction of the vascular centerline is performed on the Voronoi plot of the three-dimensional vascular model, which defines the center of the largest inset sphere. The vascular centerline is determined as a path defined on the Voronoi plot that minimizes the integral of the radius of the largest inset sphere along the path, equivalent to finding the shortest path in the radius metric.

[0049] Furthermore, in the embodiments of this application, the extraction of the vessel centerline is an important step in vascular morphological analysis. Common methods for extracting the vessel centerline may include, but are not limited to: threshold-based and filter-based methods; morphology-based methods; edge detection-based methods; watershed transform-based methods; deep learning-based methods, etc., and this application does not impose specific limitations.

[0050] As one possible implementation method, embodiments of this application can extract at least one initial path of the target blood vessel based on a three-dimensional blood vessel model, thereby obtaining at least one blood vessel centerline of the target blood vessel, and using at least one blood vessel centerline to generate a blood vessel tree, and calculate the external cross section of the lumen.

[0051] For example, embodiments of this application can extract the vascular tree and external cross-section of the target blood vessel based on a bidirectional minimum path propagation method. The main process is as follows: Figure 3 As shown, it can be:

[0052] Step S301: Extract the initial path of at least one target blood vessel and optimize it.

[0053] In this embodiment, an initial path of a target blood vessel can be extracted, and the initial path can be optimized based on path curvature, Laplace filtering, and Dijkstra's algorithm to obtain an optimal path. Other algorithms can also be selected for optimization, and this application does not impose specific limitations.

[0054] Step S302: Extract at least one blood vessel centerline.

[0055] In this application embodiment, the optimal path in step S301 can be dynamically interpolated using an iterative optimization algorithm based on cubic spline interpolation to obtain a blood vessel centerline. Alternatively, other algorithms can be used to obtain the blood vessel centerline. This application does not impose any specific limitations.

[0056] Step S303: Generate several blood vessel center lines.

[0057] In this embodiment, steps S301 and S302 can be repeated several times to obtain several blood vessel centerlines.

[0058] Step S304: Generate the vascular tree of the target blood vessel.

[0059] In this embodiment, several blood vessel centerlines can be combined into a blood vessel tree to obtain the blood vessel tree of the target blood vessel.

[0060] Step S305: Generate a more accurate vascular tree of the target blood vessel.

[0061] In this embodiment, an evolutionary method based on the Open-Snake model can be used to optimize each vessel centerline in the vascular tree to obtain a more accurate vascular tree. Other algorithms can also be used for generation, and this application does not impose specific limitations. Open-Snake is a dynamic contour model for extracting centerlines, driven by two external forces: gradient vector flow and adaptive tensile forces acting at both ends of the open curve.

[0062] Step S306: Calculate the cross-sectional area outside the lumen of the target blood vessel.

[0063] In this embodiment, a series of external lumens can be generated along the centerline of the blood vessel through the two-dimensional cross-section of the blood vessel.

[0064] Furthermore, such as Figure 4 As shown, the embodiments of this application can automatically extract the vascular tree and external cross-section of the target blood vessel based on a three-dimensional blood vessel model using a bidirectional minimum path propagation method.

[0065] In step S103, based on the outer cross section of the lumen, the vascular tree is fitted with cylinders and spheres to obtain the vascular surface fitting results. The vascular surface fitting results are used to establish the tree structure of the vascular tree, so as to obtain the sparse expression of the three-dimensional vascular model based on the tree structure.

[0066] Those skilled in the art will understand that the embodiments of this application can fit the surface of the vascular tree of a cylinder and a sphere based on the outer cross section of the lumen, and establish a tree-like structure of the vascular tree to obtain a sparse representation of a three-dimensional vascular model.

[0067] Optionally, in one embodiment of this application, the vascular tree is fitted with cylinders and spheres based on the outer cross section of the lumen to obtain the vascular surface fitting result, including: discretizing the vascular tree to obtain at least one cylinder in the vascular tree; connecting the at least one cylinder with at least one sphere based on the outer cross section of the lumen to obtain the cylinder expression and sphere expression of the vascular tree, and generating the vascular surface fitting result.

[0068] It is understood that common methods for fitting cylinders and spheres in three-dimensional data may include, but are not limited to, least squares methods, RANSAC algorithms, geometric feature-based fitting, iterative nearest point algorithms, deep learning methods, etc., and this application does not impose specific limitations.

[0069] For example, this application embodiment uses the RANSAC algorithm to fit cylinders and spheres of a blood vessel tree. The RANSAC algorithm is an iterative method used to estimate the parameters of a mathematical model from a dataset containing outliers. Specifically, the algorithm constructs the corresponding shape voxel by randomly selecting the fewest target points that can determine a basic shape voxel, and uses least squares and a voting mechanism to extract the most suitable shape voxel.

[0070] Furthermore, in this embodiment, the RANSAC algorithm using cylinders can fit cylinders of different diameters, and the cylinders are connected by spheres. By using the RANSAC algorithm to search for voxels of cylinders and spheres—the two basic shapes—in the 3D point cloud data and connecting them, the fitting result of the blood vessel surface is obtained. Specific steps are as follows: Figure 5 As shown, it can be:

[0071] Step S501: Identify a cylinder in the vascular tree.

[0072] In other words, in this embodiment of the application, three points can be randomly selected to determine a cylinder and obtain its parameters.

[0073] Step S502: Determine the points in the vascular tree that belong to this cylinder.

[0074] In other words, the embodiments of this application can calculate which points in the point cloud data belong to the cylinder according to a certain threshold. This certain threshold can be set by those skilled in the art based on actual circumstances, and this application does not impose specific limitations.

[0075] Step S503: Voting.

[0076] In other words, the embodiments of this application can use the remaining points to vote on the cylinder and record the number of votes.

[0077] Step S504: Obtain the final cylinder.

[0078] In other words, the embodiments of this application can repeat the above steps N times, compare the results obtained each time, and select the cylinder with the highest number of votes and the smallest error as the final result.

[0079] Step S505: Form all the cylinders.

[0080] In other words, in this embodiment of the application, after extracting the points belonging to the cylinder, the above steps are repeated for the remaining points until all cylinders that need to be fitted are found.

[0081] Step S506: Obtain the blood vessel surface fitting results.

[0082] In other words, the embodiments of this application can use the RANSAC algorithm to fit the remaining points between every two adjacent cylinders into sphere models of different diameters.

[0083] In some embodiments, the present application embodiments can obtain the cylindrical expression and spherical expression of the vascular tree by discretizing the vascular tree into at least one cylinder and connecting different cylinders with spheres, thereby obtaining the vascular surface fitting result of the vascular tree.

[0084] For example, embodiments of this application can be implemented according to... Figure 5 The process shown uses the RANSAC algorithm to discretize the 3D blood vessel model into a series of cylinders of different diameters, connected by spheres. The final result is as follows. Figure 6 As shown.

[0085] Optionally, in one embodiment of this application, a tree structure of a vascular tree is established using the vascular surface fitting results to obtain a sparse representation of a three-dimensional vascular model based on the tree structure. This includes: establishing a tree structure of a vascular tree based on the nodes and edges of the three-dimensional vascular model using the vascular surface fitting results; and generating vascular information for the sparse representation of the three-dimensional vascular model using the cylindrical and spherical representations of the tree structure.

[0086] It is understood that the data structure of the vascular tree in this application embodiment can be represented using a tree structure, resulting in a tree-like structure of the vascular tree. In this tree structure, each vascular branch can be represented as a node, and the branching relationships between vascular branches can be represented as edges. Specifically, the nodes in the tree structure can be divided into two categories: root nodes and non-root nodes. For non-root nodes, they all have a parent node, so the edge between them and their parent node is their incoming edge; similarly, each non-leaf node has one or more child nodes, so the edge between them and each child node is their outgoing edge. For the root node, it has no parent node, therefore it has no incoming edges; the edges connected to the root node are its outgoing edges. In this way, we can use the definitions of nodes and edges to represent the tree structure, facilitating its analysis and manipulation.

[0087] For example, this application embodiment uses a coronary artery vascular tree as an example, which is divided into two parts: the left main coronary artery and the right main coronary artery. Each main coronary artery is further divided into different branches. Therefore, in the tree structure, the entire coronary artery system can be regarded as a tree, and each node represents a blood vessel. The lines between nodes represent the connection relationship between blood vessels. In the tree structure, except for the root node, each node has a parent node, and the edge connected to that node is the edge between that node and its parent node. Figure 7 As shown, in this embodiment of the vascular tree, one edge node corresponds to one vascular branch. Within each edge node, fitted cylindrical nodes and spherical nodes are stored sequentially according to the direction of the vascular branch.

[0088] The working principle of the sparse representation method for the blood vessel model proposed in this application will be described in detail below with reference to a specific embodiment.

[0089] in, Figure 8 This is a flowchart illustrating the working principle of a sparse representation method for a blood vessel model provided according to an embodiment of this application.

[0090] Step S801: Construct a three-dimensional vascular model of the target blood vessel based on the CTA image.

[0091] In other words, embodiments of this application can reconstruct a three-dimensional vascular model of a target blood vessel from CTA images using a level set algorithm, specifically as follows: Figure 2 As shown.

[0092] Step S802: Extract the vascular tree and external cross-section of the target blood vessel.

[0093] In other words, the embodiments of this application can extract the vascular tree and external cross-section of the target blood vessel based on the bidirectional minimum path propagation method. The main process is as follows: Figure 3 As shown, further, as Figure 4 As shown in the figure, this embodiment of the application uses a bidirectional minimum path propagation method to automatically extract the vascular tree and the outer cross section of the target blood vessel.

[0094] Step S803: Fitting blood vessel surfaces based on cylinders and spheres.

[0095] In other words, the embodiments of this application can be implemented according to Figure 5 The process involves using the RANSAC algorithm to fit cylinders and spheres of the vascular tree, resulting in the following vascular surface fitting results: Figure 6 As shown.

[0096] Step S804: Establish the tree structure of the blood vessel tree.

[0097] In other words, the tree structure of the vascular tree that can be established in the embodiments of this application is as follows: Figure 7 As shown.

[0098] The sparse representation method for blood vessel models proposed in this application can extract the vascular tree and external cross-section of the target blood vessel from its constructed 3D vascular model. Based on the external cross-section, cylindrical and spherical fitting of the vascular tree is achieved, and the fitting results are used to establish a tree-like structure of the vascular tree, thus completing the sparse representation of the 3D vascular model. This reduces the computational load and complexity in virtual vascular surgery simulation, and improves the efficiency of physical simulation and computational calculation of blood vessels. Therefore, it solves the problems in related technologies where 3D vascular models are mostly represented by polygonal meshes, making real-time physical simulation difficult in dense collision detection scenarios, resulting in poor real-time performance and failing to meet practical application requirements.

[0099] Next, the sparse expression device for the blood vessel model proposed according to the embodiments of this application is described with reference to the accompanying drawings.

[0100] Figure 9 This is a block diagram of a sparse expression device for a blood vessel model provided according to an embodiment of this application.

[0101] like Figure 9 As shown, the sparse expression device 90 of the blood vessel model includes: a construction module 100, an extraction module 200, and a generation module 300.

[0102] Among them, the construction module 100 is used to construct a three-dimensional vascular model of the target blood vessel.

[0103] Extraction module 200 is used to extract the vascular tree and external cross-section of the target blood vessel based on the three-dimensional vascular model.

[0104] The generation module 300 is used to fit the vascular tree to cylinders and spheres based on the outer cross section of the lumen to obtain the vascular surface fitting results. The vascular surface fitting results are used to establish the tree structure of the vascular tree to obtain the sparse expression of the three-dimensional vascular model based on the tree structure.

[0105] Optionally, in one embodiment of this application, the construction module 100 includes: an acquisition unit, a first generation unit, and a construction unit.

[0106] The acquisition unit is used to acquire the initial angiographic image of the target blood vessel.

[0107] The first generation unit is used to segment the initial angiography image to obtain the target angiography image.

[0108] The building unit is used to extract the vascular structure from the angiography image of the target blood vessel and reconstruct a three-dimensional vascular model of the target blood vessel based on the vascular results.

[0109] Optionally, in one embodiment of this application, the extraction module 200 includes: an extraction unit, a second generation unit, and a third generation unit.

[0110] The extraction unit is used to extract at least one initial path of the target blood vessel based on the three-dimensional blood vessel model.

[0111] The second generation unit is used to obtain at least one vessel centerline of the target vessel based on at least one initial path.

[0112] The third generation unit is used to generate a vascular tree using at least one vascular centerline and calculate the external cross-section of the lumen.

[0113] Optionally, in one embodiment of this application, the generation module 300 includes: a discrete unit and a fourth generation unit.

[0114] The discrete unit is used to discretize the vascular tree to obtain at least one cylinder in the vascular tree.

[0115] The fourth generation unit is used to connect at least one cylinder with at least one sphere based on the outer cross section of the lumen to obtain the cylinder expression and sphere expression of the vascular tree, and generate the vascular surface fitting result.

[0116] Optionally, in one embodiment of this application, the generation module 300 includes: a creation unit and a fifth generation unit.

[0117] Among them, the establishment unit is used to establish the tree structure of the blood vessel tree based on the nodes and edges of the three-dimensional blood vessel model and the fitting results of the blood vessel surface.

[0118] The fifth generation unit is used to generate sparse vascular information for 3D vascular models by utilizing tree-structured cylindrical and spherical representations.

[0119] It should be noted that the foregoing explanation of the sparse expression method embodiment for the vascular model also applies to the sparse expression device for the vascular model in this embodiment, and will not be repeated here.

[0120] The sparse representation device for blood vessel models proposed in this application can extract the vascular tree and the outer cross section of the lumen from the constructed three-dimensional vascular model of the target blood vessel. Based on the outer cross section, it can fit the vascular tree to cylinders and spheres, and then use the fitting results to establish a tree structure of the vascular tree, thus completing the sparse representation of the three-dimensional vascular model. This reduces the computational load and complexity in virtual vascular surgery simulation, and improves the efficiency of physical simulation and computational calculation of blood vessels. Therefore, it solves the problems in related technologies where three-dimensional vascular models are mostly represented by polygonal meshes, making real-time physical simulation difficult in dense collision detection scenarios, resulting in poor real-time performance and failing to meet practical application requirements.

[0121] Figure 10 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. The electronic device may include:

[0122] The memory 1001, the processor 1002, and the computer program stored on the memory 1001 and capable of running on the processor 1002.

[0123] When the processor 1002 executes the program, it implements the sparse representation method of the blood vessel model provided in the above embodiments.

[0124] Furthermore, electronic devices also include:

[0125] Communication interface 1003 is used for communication between memory 1001 and processor 1002.

[0126] The memory 1001 is used to store computer programs that can run on the processor 1002.

[0127] The memory 1001 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0128] If the memory 1001, processor 1002, and communication interface 1003 are implemented independently, then the communication interface 1003, memory 1001, and processor 1002 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 10 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0129] Optionally, in a specific implementation, if the memory 1001, processor 1002, and communication interface 1003 are integrated on a single chip, then the memory 1001, processor 1002, and communication interface 1003 can communicate with each other through an internal interface.

[0130] The processor 1002 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.

[0131] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the sparse representation method of the blood vessel model described above.

[0132] This application also provides a computer program product, including a computer program that, when executed, implements the sparse representation method of the vascular model as described above.

[0133] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0134] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0135] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0136] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0137] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or more of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0138] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0139] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0140] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.

Claims

1. A sparse representation method for a blood vessel model, characterized in that, Includes the following steps: Construct a three-dimensional vascular model of the target blood vessel; Based on the three-dimensional blood vessel model, the vascular tree and external cross-section of the target blood vessel are extracted using a bidirectional minimum path propagation method. Based on the outer cross section of the lumen, the RANSAC algorithm is used to fit the vascular tree to cylinders and spheres to obtain the vascular surface fitting results. The vascular surface fitting results are used to establish the tree structure of the vascular tree to obtain the sparse representation of the three-dimensional vascular model based on the tree structure. Based on the aforementioned three-dimensional vascular model, the vascular tree and external cross-section of the target vascular vessel are extracted using a bidirectional minimum path propagation method, including: An initial path for a target blood vessel is extracted, and the initial path is optimized based on path curvature, Laplace filtering, and Dijkstra's algorithm to obtain an optimal path. An iterative optimization algorithm based on cubic spline interpolation is used to dynamically interpolate the optimal path to obtain a blood vessel centerline. An evolutionary method based on the Open-Snake model optimizes the vascular centerline in the vascular tree to generate the vascular tree; Along the centerline of the blood vessel, the outer cross section of the lumen is generated through the two-dimensional cross-section of the blood vessel; Based on the external cross-section of the lumen, the RANSAC algorithm is used to fit the vascular tree to cylinders and spheres, including: Discretize the vascular tree and randomly select three points to determine a cylinder; Based on a certain threshold, calculate which points in the point cloud data belong to this cylinder; Use the remaining points to vote on the cylinder and record the number of votes. Repeat the above steps, compare the results obtained each time, and select the cylinder with the highest number of votes and the smallest error as the final result; For the remaining points between every two adjacent cylinders, the RANSAC algorithm is used to fit a sphere model with different diameters to obtain the fitting results of the blood vessel surface.

2. The sparse representation method for the vascular model according to claim 1, characterized in that, The construction of the three-dimensional vascular model of the target blood vessel includes: Obtain an initial angiographic image of the target blood vessel; The initial angiography image is segmented to obtain the target angiography image; Extract the vascular structure from the angiographic image of the target blood vessel, and reconstruct the three-dimensional vascular model of the target blood vessel based on the vascular structure.

3. The sparse representation method for the vascular model according to claim 1, characterized in that, The step of establishing a tree-like structure of the vascular tree using the vascular surface fitting results, and obtaining a sparse representation of the three-dimensional vascular model based on the tree-like structure, includes: Based on the nodes and edges of the three-dimensional blood vessel model, the tree structure of the blood vessel tree is established using the fitting results of the blood vessel surface; The tree-structured cylindrical and spherical representations are used to generate sparse vascular information for the three-dimensional vascular model.

4. A sparse expression device for a vascular model, used to implement the sparse expression method for a vascular model as described in any one of claims 1-3, characterized in that, include: The building module is used to construct a three-dimensional vascular model of the target blood vessel; The extraction module is used to extract the vascular tree and external cross-section of the target blood vessel based on the three-dimensional vascular model; The generation module is used to fit the vascular tree to cylinders and spheres based on the outer cross section of the lumen to obtain the vascular surface fitting result, and to establish the tree structure of the vascular tree based on the vascular surface fitting result to obtain the sparse representation of the three-dimensional vascular model based on the tree structure.

5. The sparse expression device for a vascular model according to claim 4, characterized in that, The building module includes: The acquisition unit is used to acquire the initial angiographic image of the target blood vessel; The first generation unit is used to segment the initial angiography image to obtain the target angiography image; A construction unit is used to extract the vascular structure from the target angiography image and reconstruct the three-dimensional vascular model of the target blood vessel based on the vascular structure.

6. The sparse expression device for a vascular model according to claim 4, characterized in that, The extraction module includes: An extraction unit is used to extract at least one initial path of the target blood vessel based on the three-dimensional blood vessel model; The second generation unit is used to obtain at least one vessel centerline of the target vessel based on the at least one initial path; The third generation unit is used to generate the vascular tree using the at least one vascular centerline and to calculate the external cross-section of the lumen.

7. The sparse expression device for a vascular model according to claim 4, characterized in that, The generation module includes: Discretizable unit, used to discretize the vascular tree to obtain at least one cylinder in the vascular tree; The fourth generation unit is used to connect the at least one cylinder with at least one sphere based on the outer cross section of the lumen to obtain the cylinder expression and sphere expression of the vascular tree, and generate the vascular surface fitting result.

8. The sparse expression device for a vascular model according to claim 4, characterized in that, The generation module includes: A building unit is used to build the tree structure of the blood vessel tree based on the nodes and edges of the three-dimensional blood vessel model and the fitting results of the blood vessel surface; The fifth generation unit is used to generate vascular information for the sparse representation of the three-dimensional vascular model using the cylindrical and spherical representations of the tree structure.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the sparse representation method of the vascular model as described in any one of claims 1-3.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the sparse representation method of the vascular model as described in any one of claims 1-3.

11. A computer program product, characterized in that, Includes a computer program, which, when executed, is used to implement the sparse representation method of the vascular model as described in any one of claims 1-3.

Images