A blood vessel bifurcation level judgment method and device and a storage medium

Abstract

The present application relates to a kind of vascular bifurcation level judgment method, device and storage medium, applied to the technical field of imageology, comprising: by obtaining skeleton from blood vessel original image, then by the skeleton point in skeleton, minimum spanning tree is generated using graph algorithm, all points of minimum spanning tree are given label value, since the node of minimum spanning tree and the skeleton point of blood vessel skeleton one-to-one correspondence, after obtaining the label value of the node of minimum spanning tree, the label value of skeleton point is obtained, then the skeleton point closest to each voxel point in blood vessel original image is obtained, the label value of this skeleton point is given to corresponding voxel point, the judgment of vascular bifurcation level is completed, by the skeleton obtained, the original image can be greatly simplified and the topological structure of original image is retained more completely, and graph algorithm is a relatively effective algorithm for solving spatial topology problem, and the combination of the two can achieve the purpose of vascular bifurcation judgment, effectively improve the judgment efficiency.

Description

Technical Field

[0001] This invention relates to the field of imaging research technology, specifically to a method, device, and storage medium for determining the level of blood vessel bifurcation. Background Technology

[0002] In imaging studies such as CT and MR, it is sometimes necessary to determine the bifurcation level of vessels such as the portal vein in the liver for preoperative assessment or other clinical research. In current technology, manual annotation is usually used. However, relying solely on manual annotation is not only inefficient but also consumes a lot of human resources. Therefore, it is necessary to seek an automated auxiliary judgment method. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method, device and storage medium for determining the level of blood vessel bifurcation, so as to solve the problem that in the prior art, the determination of the level of blood vessel bifurcation usually adopts the form of manual annotation, which is not only inefficient but also consumes human resources.

[0004] According to a first aspect of the present invention, a method for determining the level of blood vessel bifurcation is provided, the method comprising:

[0005] The original blood vessel image is preprocessed to obtain the preprocessed blood vessel image;

[0006] The vascular skeleton is obtained based on all voxel points of the preprocessed vascular image.

[0007] Generate a minimum spanning tree from all the skeleton points in the vascular skeleton using a graph algorithm;

[0008] Starting from the root of the minimum spanning tree, assign a label value to the starting point. Then, use a depth-first search algorithm to traverse all nodes in the minimum spanning tree. When a branch point is reached, change the label value of the subsequent nodes until the label values ​​of all nodes in the minimum spanning tree are assigned.

[0009] The label values ​​of the skeleton points of the blood vessel skeleton are obtained based on the label values ​​of the nodes in the minimum spanning tree. The skeleton point that is closest to each voxel point in the original blood vessel image in the blood vessel skeleton is obtained, and the label value of the skeleton point is assigned to the corresponding voxel point to complete the blood vessel bifurcation and classification of the original blood vessel image.

[0010] Preferably,

[0011] The preprocessing of the original blood vessel image to obtain the preprocessed blood vessel image includes:

[0012] Extract the largest connected component from the original blood vessel image, and then resample the labeled blood vessel image with the largest connected component at equal intervals in all directions;

[0013] The original blood vessel image after equidistant resampling is surface smoothed to obtain the preprocessed blood vessel image.

[0014] Preferably,

[0015] The step of generating a minimum spanning tree from all skeletal points in the vascular skeleton using a graph algorithm includes:

[0016] Obtain all skeleton points of the vascular skeleton, and generate the adjacency matrix of the vascular skeleton based on the spatial distance between any two skeleton points;

[0017] An undirected weighted graph is generated based on the adjacency matrix of the vascular skeleton;

[0018] Obtain the minimum spanning tree of the undirected weighted graph.

[0019] Preferably,

[0020] The step of obtaining all skeletal points of the vascular skeleton and generating an adjacency matrix of the vascular skeleton based on the spatial distance between any two skeletal points includes:

[0021] Obtain the N skeleton points contained in the vascular skeleton, each skeleton point including its spatial coordinates, generate an N-order square matrix composed of the distances between each skeleton point, and use this square matrix as the adjacency matrix.

[0022] Preferably,

[0023] The generation of the undirected weighted graph based on the adjacency matrix of the vascular skeleton includes:

[0024] Connect each skeleton point in the vascular skeleton to all other skeleton points to obtain a fully connected graph;

[0025] Select any two skeleton points in the fully connected graph, and weight the connection strength of the two skeleton points according to the distance in the adjacency matrix to obtain an undirected weighted graph.

[0026] Preferably,

[0027] The process begins by assigning a label value to the root of the minimum spanning tree. Starting from this root, a depth-first search algorithm is used to traverse all nodes in the minimum spanning tree. When a branch point is reached, the label values ​​of the nodes following that branch point are changed. This process continues until all nodes in the minimum spanning tree have been labeled.

[0028] Starting from the initial point, obtain the degree of each node. The degree of a node is the sum of the number of nodes directly connected to it. After the initial point, if the degree of each node traversed is less than or equal to 2, it means that the node is not a branch point. Assign the initial point's label value to the node. Continue until a node with a degree greater than or equal to 3 is encountered. This node is a branch point. For each branch point, assign the initial point's label value. Assign new label values ​​to the nodes after the branch point. Continue until a new branch point is encountered. Assign new label values ​​to the nodes after the new branch point again. Continue assigning values ​​to all nodes.

[0029] According to a second aspect of the present invention, a device for determining the level of blood vessel bifurcation is provided, the device comprising:

[0030] Tag image acquisition module: used to preprocess the original blood vessel image to obtain the preprocessed blood vessel image;

[0031] Skeleton acquisition module: used to acquire the vascular skeleton based on all voxel points of the preprocessed vascular image;

[0032] Minimum Spanning Tree Generation Module: Used to generate a minimum spanning tree from all skeleton points in the vascular skeleton using a graph algorithm;

[0033] The assignment module is used to assign a label value to the root of the minimum spanning tree. Starting from the root, it traverses all nodes in the minimum spanning tree using a depth-first search algorithm. When a branch point is reached, the label value of the subsequent nodes is changed until the label values ​​of all nodes in the minimum spanning tree are assigned.

[0034] The blood vessel bifurcation level determination module is used to obtain the label value of the skeleton point of the blood vessel skeleton based on the label value of the node in the minimum spanning tree, obtain the skeleton point that is closest to each voxel point in the original blood vessel image in the blood vessel skeleton, and assign the label value of the skeleton point to the corresponding voxel point to complete the blood vessel bifurcation level determination of the original blood vessel image.

[0035] According to a third aspect of the present invention, a storage medium is provided, the storage medium storing a computer program, which, when executed by a host controller, implements the steps of the above-described method.

[0036] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0037] In this application, an image skeleton acquisition algorithm is used to obtain the skeleton from the blood vessel label image. Then, a minimum spanning tree is generated using a graph algorithm for all skeleton points in the blood vessel skeleton. Label values ​​are assigned to all points in the minimum spanning tree. Since the nodes of the minimum spanning tree correspond one-to-one with the skeleton points of the blood vessel skeleton, the label values ​​of the skeleton points are obtained after obtaining the label values ​​of the nodes of the minimum spanning tree. Then, the skeleton point closest to each voxel point in the original image is obtained, and the label value of the skeleton point is assigned to the corresponding voxel point to complete the determination of blood vessel bifurcation level. In this application, the obtained skeleton can greatly simplify the original image and retain the topological structure of the original image relatively completely. Graph algorithms are relatively effective algorithms for solving spatial topology problems. The combination of the two can achieve the purpose of blood vessel bifurcation determination without manual intervention, effectively improving the efficiency of blood vessel bifurcation level determination.

[0038] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0039] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0040] Figure 1 This is a flowchart illustrating a method for determining the level of blood vessel bifurcation according to an exemplary embodiment;

[0041] Figure 2 This is a schematic diagram of a portal vein label without determining the bifurcation level, according to an exemplary embodiment.

[0042] Figure 3 This is a schematic diagram of a skeleton according to an exemplary embodiment;

[0043] Figure 4 This is a schematic diagram of an undirected weighted graph according to an exemplary embodiment;

[0044] Figure 5 This is a schematic diagram of a minimum spanning tree according to an exemplary embodiment;

[0045] Figure 6 This is a schematic diagram illustrating the assignment of minimum spanning tree label values ​​according to an exemplary embodiment;

[0046] Figure 7 This is a schematic diagram of skeleton staining according to an exemplary embodiment;

[0047] Figure 8 This is a schematic diagram illustrating the skeleton point for finding the skeleton closest to the original tag voxel point according to an exemplary embodiment.

[0048] Figure 9 This is a schematic diagram of portal vein label vessel bifurcation staining according to an exemplary embodiment;

[0049] Figure 10 This is a system schematic diagram of a blood vessel bifurcation level determination device according to an exemplary embodiment;

[0050] In the attached diagram: 1-Label image acquisition module, 2-Skeleton acquisition module, 3-Minimum spanning tree generation module, 4-Assignment module, 5-Vascular bifurcation level judgment module. Detailed Implementation

[0051] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0052] Example 1

[0053] Figure 1 This is a flowchart illustrating a method for determining the level of blood vessel bifurcation according to an exemplary embodiment, such as... Figure 1 As shown, the method includes:

[0054] S1, preprocess the original blood vessel image to obtain the preprocessed blood vessel image;

[0055] S2, Obtain the vascular skeleton based on all voxel points of the preprocessed vascular image;

[0056] S3, use a graph algorithm to generate a minimum spanning tree from all the skeleton points in the vascular skeleton;

[0057] S4. Starting from the root of the minimum spanning tree, assign a label value to the starting point. Starting from the starting point, traverse all nodes in the minimum spanning tree using a depth-first search algorithm. When a branch point is reached, change the label value of the subsequent nodes until the label values ​​of all nodes in the minimum spanning tree are assigned.

[0058] S5. Obtain the label value of the skeleton point of the blood vessel skeleton according to the label value of the node in the minimum spanning tree, obtain the skeleton point that is closest to each voxel point in the original blood vessel image in the blood vessel skeleton, and assign the label value of the skeleton point to the corresponding voxel point to complete the blood vessel bifurcation and classification of the original blood vessel image.

[0059] It is understood that this embodiment uses a liver portal vein label as an example, as shown in the attached document. Figure 2As shown, before obtaining the skeleton, some preprocessing operations are performed on the blood vessel labels to obtain a preprocessed blood vessel image. Currently, many open-source computer graphics algorithms include algorithms for obtaining image skeletons, such as the `skeleton_3d` function in the `scikit-image` library, which can obtain the skeleton of a 3D graphic, as shown in the attached image. Figure 3 As shown, the skeleton is obtained from voxel points of the preprocessed blood vessel image. It's worth emphasizing that a voxel is a concept in 3D graphics. In a 2D image, the smallest display unit is called a 'pixel', and in 3D graphics, the smallest display unit is called a 'voxel'. Just as countless pixels make up an image, countless voxels make up a label graphic in 3D space. Obtaining voxels is as simple as obtaining pixels; given the coordinates, the voxel points are determined. After obtaining the skeleton, the minimum spanning tree of all skeleton points is used, as shown in the attached diagram. Figure 5 As shown, starting from the root of the minimum spanning tree (the root can be manually marked or located using existing labels), assign a label value to the starting point. Starting from the starting point, a depth-first search algorithm is used to traverse all points in the minimum spanning tree. When a branch point is reached, the label values ​​of subsequent points are changed until the label values ​​of all points in the minimum spanning tree are assigned, as shown in the attached diagram. Figure 6 As shown, after obtaining the label values ​​of the nodes in the minimum spanning tree, the label values ​​of all skeletal points in the vascular skeleton are obtained. Based on the label values ​​of the points in the minimum spanning tree, the skeleton is colored according to different label values, as shown in the attached diagram. Figure 7 As shown, all voxel points in the original blood vessel image are obtained, and the closest skeleton point in the skeleton for each voxel point is obtained, as shown in the attached figure. Figure 8 As shown, the label value of the skeleton point is assigned to the corresponding voxel point to complete the vascular bifurcation classification of the original vascular image. To facilitate a more intuitive view of the classification results, this embodiment will also color the skeleton according to different label values ​​of the points in the minimum spanning tree, as shown in the attached figure. Figure 7 As shown in the attached figure, the original blood vessel image is stained according to different label values ​​based on the voxel point label values. Figure 9 As shown, it is worth emphasizing that the appendix Figure 8 In this example, assuming A, B, and C are voxel points in the original blood vessel image, to determine their appropriate label values, we only need to determine the label value of the skeleton point closest to them in the skeleton. For example, the skeleton point closest to A is D, and since D's label value is already assigned to 1, A is also assigned a label value of 1. The skeleton point closest to B is E, and since E's label value is 2, B should also have a label value of 2. Similarly, C should also have a label value of 2. Thus, based on the label values ​​of the voxel points in the original blood vessel image, the original blood vessel image is stained, thereby obtaining the attached... Figure 9In this application, an image skeleton acquisition algorithm is used to obtain the skeleton from the blood vessel label image. Then, an adjacency matrix is ​​generated based on the spatial distance between any two voxel points in the skeleton. A minimum spanning tree is generated using the adjacency matrix. A graph algorithm is used to assign label values ​​to all points in the minimum spanning tree. The skeleton is then colored based on the label values. Finally, the blood vessel label is colored based on the skeleton coloring, thus completing the determination of the blood vessel bifurcation level. In this application, an image skeleton acquisition algorithm is used to obtain the skeleton from the blood vessel label image. Then, a minimum spanning tree is generated using a graph algorithm for all skeleton points in the blood vessel skeleton, and label values ​​are assigned to all points in the minimum spanning tree. Since the nodes of the minimum spanning tree correspond one-to-one with the skeleton points of the vascular skeleton, after obtaining the label values ​​of the nodes of the minimum spanning tree, the label values ​​of the skeleton points of the skeleton are obtained. Then, the skeleton point closest to each voxel point in the original image is obtained, and the label value of the skeleton point is assigned to the corresponding voxel point to complete the judgment of the vascular bifurcation level. In this application, the obtained skeleton can greatly simplify the original image and retain the topological structure of the original image relatively completely. Graph algorithms are relatively effective algorithms for solving spatial topology problems. The combination of the two can achieve the purpose of vascular bifurcation judgment without manual intervention, effectively improving the efficiency of vascular bifurcation level judgment.

[0060] It is worth emphasizing that, for ease of understanding, the label value of each node corresponds one-to-one with the label value of the skeleton point in the vascular skeleton, and the voxel point in the vascular skeleton is defined as the skeleton point.

[0061] Preferably,

[0062] The preprocessing of the original blood vessel image to obtain the preprocessed blood vessel image includes:

[0063] Extract the largest connected component from the original blood vessel image, and then resample the labeled blood vessel image with the largest connected component at equal intervals in all directions;

[0064] A surface smoothing operation is performed on the original blood vessel image after equidistant resampling to obtain a preprocessed blood vessel image;

[0065] Understandably, the process involves extracting the largest connected component of the image. Since there is sometimes noise and detached labels around the image, to remove this interference, only the image containing the largest connected component is processed. The image is then resampled at equal intervals in all directions. In most cases, the sampling intervals of the labels in the X, Y, and Z directions are different, which can lead to deviations in the obtained skeleton. To maintain directional consistency and obtain more accurate skeleton points, equal-interval resampling is performed on the image in all directions. For smoothing the image surface, some labels may have poor quality, resulting in uneven surfaces. Therefore, depending on the actual situation, closing operations (used when the surface has small grooves) or erosion operations (used when the surface has small bumps) can be used to smooth the surface. For the closing operation, we chose the `binary_closing` function from scikit-image; for the erosion operation, we chose the `binary_erosion` function from scikit-image.

[0066] Preferably,

[0067] The step of generating a minimum spanning tree from all skeletal points in the vascular skeleton using a graph algorithm includes:

[0068] Obtain all skeleton points of the vascular skeleton, and generate the adjacency matrix of the vascular skeleton based on the spatial distance between any two skeleton points;

[0069] An undirected weighted graph is generated based on the adjacency matrix of the vascular skeleton;

[0070] Obtain the minimum spanning tree of the undirected weighted graph;

[0071] Understandably, the adjacency matrix of the vascular skeleton is generated based on the spatial distance between any two skeleton points, and an undirected weighted graph of the skeleton is generated based on the adjacency matrix, as shown in the attached figure. Figure 4 As shown, obtain the minimum spanning tree of this undirected weighted graph, as attached. Figure 5 As shown, an undirected weighted fully connected graph has a drawback: all points are related to each other, which undoubtedly increases the complexity of the analysis. In the skeleton points, we only need to focus on points that are spatially close. For example, we only need to focus on the connection between point A and point B, and do not need to focus on the connection between point A and point E. Therefore, we simplify the fully connected graph by only obtaining its minimum spanning tree. The minimum spanning tree uses the spanning_tree function in igraph.

[0072] It is worth emphasizing that the methods for generating minimum spanning trees are well-established existing techniques, such as Kruskal's algorithm: edges are added to the spanning tree in ascending order of their weights, but if adding an edge would create a cycle with the spanning tree, then that edge is not added, until the tree contains all vertices; or Prim's algorithm: starting from any tree containing a single vertex, vertices are added to it one by one with the smallest weight, until all vertices have been added.

[0073] Preferably,

[0074] The step of obtaining all skeletal points of the vascular skeleton and generating an adjacency matrix of the vascular skeleton based on the spatial distance between any two skeletal points includes:

[0075] Obtain the N skeleton points contained in the vascular skeleton, each skeleton point includes its spatial coordinates, generate an N-order square matrix composed of the distances between each skeleton point, and use this square matrix as the adjacency matrix.

[0076] Understandably, we obtain all skeleton points representing the skeleton and generate an adjacency matrix based on the distance between any two points. Assume there are N skeleton points with spatial coordinates (x0, y0, z0), (x1, y1, z1), ..., (x...). N-1 y N-1 , z N-1 Then, an N-order square matrix consisting of the distances between points can be generated, as shown below:

[0077]

[0078] in,

[0079]

[0080] In the above formula, d i，j This represents the distance between the i-th point and the j-th point.

[0081] Preferably,

[0082] The generation of the undirected weighted graph based on the adjacency matrix of the vascular skeleton includes:

[0083] Connect each skeleton point in the vascular skeleton to all other skeleton points to obtain a fully connected graph;

[0084] Select any two skeleton points in the fully connected graph, and weight the connection strength of the two skeleton points according to the distance in the adjacency matrix to obtain an undirected weighted graph.

[0085] Understandably, once the adjacency matrix is ​​calculated, it can be used to organize the original N skeleton points. Here, we generate an undirected weighted graph. A graph is a data structure, a way of organizing data. Using a graph data structure, we can more easily manage the relationships between data. In our algorithm, we use the open-source graph framework igraph (https: / / igraph.org / ) for graph generation and related calculations, as detailed below: (See attached image) Figure 4 As shown, assuming append Figure 4 The graph represents 5 points in a plane. We connect each point to the other four points to form a graph. Because there is a connection between every two points, it is called a fully connected graph. We do not care about the direction of the connection between two points, so we also call it an undirected graph. Furthermore, we not only care about the connection between two points, but also the strength of their connection. Therefore, we use the distance between two points to weight them. Combined, we get an undirected weighted graph.

[0086] Preferably,

[0087] The process begins by assigning a label value to the root of the minimum spanning tree. Starting from this root, a depth-first search algorithm is used to traverse all nodes in the minimum spanning tree. When a branch point is reached, the label values ​​of the nodes following that branch point are changed. This process continues until all nodes in the minimum spanning tree have been labeled.

[0088] Starting from the initial point, obtain the degree of each node. The degree of a node is the sum of the number of nodes directly connected to it. After the initial point, if the degree of each node traversed is less than or equal to 2, it means that the node is not a branch point. Assign the initial point's label value to the node. Continue until a node with a degree greater than or equal to 3 is encountered. This node is a branch point. Assign the initial point's label value to the branch point as well. Assign new label values ​​to the nodes after the branch point. Continue until a new branch point is encountered. Assign new label values ​​to the nodes after the new branch point again. Continue until all nodes are assigned labels.

[0089] Understandably, as shown in the attached document Figure 5As shown, assuming the starting point is A, the initial label value is 1 (different numbers are used for different branches for easy distinction, and for simplicity, we use incrementing integers starting from 1). The degree of A is 1, which represents the sum of the number of points directly connected to a given point. Since only point B is connected to A, and point B has a degree of 2, it is connected to C in addition to A. Therefore, B is not a branch, and A's label value of 1 is assigned to B as well. For point C, its degree is 3. C is connected to D, B, and E respectively. Since C's degree is 3, satisfying the condition of being greater than or equal to 3, the label value of branch point C continues from the label values ​​of points B and A, which are 1. For points D and E, new label values ​​of 2 are assigned (using incrementing integers starting from 1). If there are further branches after point D or E, the label value of the points after those branches is 3. Label values ​​are assigned to all points in the minimum spanning tree, as shown in the attached diagram. Figure 6 As shown.

[0090] Example 2

[0091] Figure 10 This is a schematic diagram of a blood vessel bifurcation level determination device according to an exemplary embodiment, including:

[0092] Label image acquisition module 1: used to preprocess the original blood vessel image to obtain the preprocessed blood vessel image;

[0093] Skeleton acquisition module 2: used to acquire the vascular skeleton based on all voxel points of the preprocessed vascular image;

[0094] Minimum Spanning Tree Generation Module 3: Used to generate a minimum spanning tree from all skeleton points in the vascular skeleton using a graph algorithm;

[0095] Assignment Module 4: It is used to assign a label value to the root of the minimum spanning tree as the starting point. Starting from the starting point, it traverses all nodes in the minimum spanning tree using a depth-first search algorithm. When a branch point is reached, the label value of the subsequent nodes of the branch point is changed until the label values ​​of all nodes in the minimum spanning tree are assigned.

[0096] Blood vessel bifurcation level judgment module 5: It is used to obtain the label value of the skeleton point of the blood vessel skeleton based on the label value of the node in the minimum spanning tree, obtain the skeleton point that is closest to each voxel point in the original blood vessel image in the blood vessel skeleton, and assign the label value of the skeleton point to the corresponding voxel point to complete the blood vessel bifurcation level of the original blood vessel image.

[0097] It is understood that this embodiment also discloses a vessel bifurcation level determination device, including: a label image acquisition module 1 for preprocessing the original vessel image to obtain a preprocessed vessel image; a skeleton acquisition module 2 for acquiring the vessel skeleton based on all voxel points of the preprocessed vessel image; a minimum spanning tree generation module 3 for generating a minimum spanning tree from all skeleton points of the vessel skeleton using a graph algorithm; an assignment module 4 for assigning a label value to the root of the minimum spanning tree as the starting point, and traversing all nodes in the minimum spanning tree using a depth-first search algorithm from the starting point. When a bifurcation point is reached, the subsequent nodes of the bifurcation point change their label values ​​until the label values ​​of all nodes in the minimum spanning tree are assigned; and a vessel bifurcation level determination module 5 for obtaining the label values ​​of the skeleton points of the vessel skeleton based on the label values ​​of the nodes in the minimum spanning tree, and obtaining the closest value of each voxel point in the original vessel image to the vessel skeleton. The skeleton points are obtained, and their label values ​​are assigned to the corresponding voxel points to complete the classification of blood vessel bifurcation in the original blood vessel image. In this application, the skeleton is obtained from the labeled blood vessel image by an image skeleton acquisition algorithm. Then, all skeleton points in the blood vessel skeleton are used to generate a minimum spanning tree using a graph algorithm. Label values ​​are assigned to all points in the minimum spanning tree. Since the nodes of the minimum spanning tree correspond one-to-one with the skeleton points of the blood vessel skeleton, the label values ​​of the skeleton points of the skeleton are obtained after obtaining the label values ​​of the nodes of the minimum spanning tree. Then, the skeleton point closest to each voxel point in the original image is obtained, and the label value of the skeleton point is assigned to the corresponding voxel point to complete the judgment of blood vessel bifurcation level. In this application, the obtained skeleton can greatly simplify the original image and retain the topological structure of the original image relatively completely. Graph algorithms are relatively effective algorithms for solving spatial topology problems. The combination of the two can achieve the purpose of blood vessel bifurcation judgment without manual intervention, effectively improving the efficiency of blood vessel bifurcation level judgment.

[0098] Example 3:

[0099] This embodiment provides a storage medium storing a computer program, which, when executed by a host controller, implements the various steps in the above method.

[0100] It is understood that the storage medium mentioned above can be a read-only memory, a hard disk, or an optical disk, etc.

[0101] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0102] It should be noted that in the description of this invention, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this invention, unless otherwise stated, "a plurality of" means at least two.

[0103] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention 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 will be understood by those skilled in the art to which embodiments of the invention pertain.

[0104] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination 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.

[0105] 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.

[0106] Furthermore, the functional units in the various embodiments of the present invention 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.

[0107] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0108] In the description of this specification, 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 the invention. 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.

[0109] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for determining the level of blood vessel bifurcation, characterized in that, The method includes: The original blood vessel image is preprocessed to obtain the preprocessed blood vessel image; The vascular skeleton is obtained based on all voxel points of the preprocessed vascular image. Generate a minimum spanning tree from all the skeleton points in the vascular skeleton using a graph algorithm; The step of generating a minimum spanning tree from all skeletal points in the vascular skeleton using a graph algorithm includes: Obtain all skeleton points of the vascular skeleton, and generate the adjacency matrix of the vascular skeleton based on the spatial distance between any two skeleton points; An undirected weighted graph is generated based on the adjacency matrix of the vascular skeleton; Obtain the minimum spanning tree of the undirected weighted graph; Starting from the root of the minimum spanning tree, assign a label value to the starting point. Then, use a depth-first search algorithm to traverse all nodes in the minimum spanning tree. When a branch point is reached, change the label value of the subsequent nodes until the label values ​​of all nodes in the minimum spanning tree are assigned. The label values ​​of the skeleton points of the blood vessel skeleton are obtained based on the label values ​​of the nodes in the minimum spanning tree. The skeleton point that is closest to each voxel point in the original blood vessel image in the blood vessel skeleton is obtained, and the label value of the skeleton point is assigned to the corresponding voxel point to complete the blood vessel bifurcation and classification of the original blood vessel image.

2. The method according to claim 1, characterized in that, The preprocessing of the original blood vessel image to obtain the preprocessed blood vessel image includes: Extract the largest connected component from the original blood vessel image, and then resample the labeled blood vessel image with the largest connected component at equal intervals in all directions; The original blood vessel image after equidistant resampling is surface smoothed to obtain the preprocessed blood vessel image.

3. The method according to claim 2, characterized in that, The step of obtaining all skeletal points of the vascular skeleton and generating an adjacency matrix of the vascular skeleton based on the spatial distance between any two skeletal points includes: Obtain the N skeleton points contained in the vascular skeleton, each skeleton point including its spatial coordinates, generate an N-order square matrix composed of the distances between each skeleton point, and use this square matrix as the adjacency matrix.

4. The method according to claim 3, characterized in that, The generation of the undirected weighted graph based on the adjacency matrix of the vascular skeleton includes: Connect each skeleton point in the vascular skeleton to all other skeleton points to obtain a fully connected graph; Select any two skeleton points in the fully connected graph, and weight the connection strength of the two skeleton points according to the distance in the adjacency matrix to obtain an undirected weighted graph.

5. The method according to claim 1, characterized in that, The process begins by assigning a label value to the root of the minimum spanning tree. Starting from this root, a depth-first search algorithm is used to traverse all nodes in the minimum spanning tree. When a branch point is reached, the label values ​​of the nodes following that branch point are changed. This process continues until all nodes in the minimum spanning tree have been labeled. Starting from the initial point, obtain the degree of each node. The degree of a node is the sum of the number of nodes directly connected to it. After the initial point, if the degree of each node traversed is less than or equal to 2, it means that the node is not a branch point. Assign the initial point's label value to the node. Continue until a node with a degree greater than or equal to 3 is encountered. This node is a branch point. For each branch point, assign the initial point's label value. Assign new label values ​​to the nodes after the branch point. Continue until a new branch point is encountered. Assign new label values ​​to the nodes after the new branch point again. Continue assigning values ​​to all nodes.

6. A device for determining the level of blood vessel bifurcation, characterized in that, The device includes: Tag image acquisition module: used to preprocess the original blood vessel image to obtain the preprocessed blood vessel image; Skeleton acquisition module: used to acquire the vascular skeleton based on all voxel points of the preprocessed vascular image; Minimum Spanning Tree Generation Module: Used to generate a minimum spanning tree from all skeleton points in the vascular skeleton using a graph algorithm; The step of generating a minimum spanning tree from all skeletal points in the vascular skeleton using a graph algorithm includes: Obtain all skeleton points of the vascular skeleton, and generate the adjacency matrix of the vascular skeleton based on the spatial distance between any two skeleton points; An undirected weighted graph is generated based on the adjacency matrix of the vascular skeleton; Obtain the minimum spanning tree of the undirected weighted graph; The assignment module is used to assign a label value to the root of the minimum spanning tree. Starting from the root, it traverses all nodes in the minimum spanning tree using a depth-first search algorithm. When a branch point is reached, the label value of the subsequent nodes is changed until the label values ​​of all nodes in the minimum spanning tree are assigned. The blood vessel bifurcation level determination module is used to obtain the label value of the skeleton point of the blood vessel skeleton based on the label value of the node in the minimum spanning tree, obtain the skeleton point that is closest to each voxel point in the original blood vessel image in the blood vessel skeleton, and assign the label value of the skeleton point to the corresponding voxel point to complete the blood vessel bifurcation level determination of the original blood vessel image.

7. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by the main controller, implements each step of the method for determining the level of blood vessel bifurcation as described in any one of claims 1-5.

Images