CN-120147355-B - Individuation partitioning method, device, medium and product based on brain surface geometric characteristics

CN120147355BCN 120147355 BCN120147355 BCN 120147355BCN-120147355-B

Abstract

The application discloses a brain surface geometric feature-based individuation partitioning method, a device, a medium and a product, and relates to the technical field of neural imaging, wherein the method comprises the steps of dividing each vertex into a sulcus seed point, a gyrus seed point and residual vertexes based on the shape index of each vertex, so as to obtain a plurality of sulcus seed initial areas and gyrus seed initial areas; combining the initial areas of the cerebral sulcus seeds and the initial areas of the cerebral gyrus seeds which are communicated respectively to obtain a plurality of combined areas, and determining the initial areas which are not communicated as independent areas; and determining all vertexes in the merging area and the independent area with the vertex number smaller than the preset number and all the rest vertexes as transition vertexes. Determining merging areas and independent areas with the number of vertexes being larger than and/or equal to the preset number as initial inclusion areas; and iteratively incorporating each transition vertex to obtain a brain personalized partition map, thereby realizing the partition of the brain. The application improves the accuracy of brain region division.

Inventors

LIU ZHAO
LUAN GUOMING
GUAN YUGUANG
ZHOU JIAN

Assignees

北京三博脑科医院有限公司

Dates

Publication Date: 20260508
Application Date: 20250314

Claims (9)

1. A brain surface geometry based personalized partitioning method, the brain surface geometry based personalized partitioning method comprising: Acquiring three-dimensional grid data of the brain surface, wherein the three-dimensional grid data comprises a plurality of vertexes and a plurality of edges connecting two adjacent vertexes; Determining a shape index of each vertex based on the three-dimensional mesh data; Dividing each vertex into a sulcus seed point, a brain return seed point and the rest vertex based on the shape index of each vertex; determining a plurality of initial areas of the sulcus seed based on the plurality of neighboring sulcus seed points, and determining a plurality of initial areas of the gyrus seed based on the plurality of neighboring gyrus seed points; determining connectivity between initial areas of the brain ditch seeds and connectivity between initial areas of the brain return seeds respectively, wherein the connectivity is connected or disconnected; Combining all the initial areas of the brain ditch seeds with connectivity as communication to obtain a plurality of brain ditch seed combining areas, and combining all the initial areas of the brain return seeds with connectivity as communication to obtain a plurality of brain return seed combining areas; determining all initial areas of the cerebral sulcus seeds which are not communicated as independent areas of the cerebral sulcus seeds, and determining all initial areas of the cerebral gyrus seeds which are not communicated as independent areas of the cerebral gyrus seeds; Determining all vertexes in the sulcus seed merging region, the sulcus seed independent region and the sulcus seed independent region with the vertex number smaller than the preset number and all the rest vertexes as transition vertexes to obtain an initial transition vertex set; Determining a cerebral sulcus seed merging region, a cerebral gyrus seed merging region, a cerebral sulcus seed independent region and a cerebral gyrus seed independent region with the vertex number being larger than and/or equal to the preset number as initial inclusion regions; iteratively incorporating each transition vertex in the initial transition vertex set based on each initial incorporating region to obtain a plurality of brain ditch seed target regions and brain return seed target regions, thereby obtaining a brain personalized partition map and realizing the partition of the brain; iteratively incorporating each transition vertex in the initial transition vertex set based on each initial incorporating region to obtain a plurality of brain ditch seed target regions and brain return seed target regions, thereby obtaining a brain personalized partition map, and realizing the partition of the brain, comprising: Based on each initial inclusion region, performing repeated iterative inclusion on all transition vertexes in the initial transition vertex set to obtain a plurality of brain ditch seed target regions and brain return seed target regions, thereby obtaining a brain personalized partition map and realizing the partition of the brain, wherein the inclusion process under any current iteration times comprises the following steps: Judging whether each transition vertex in a transition vertex set before updating under the current iteration number and any vertex in an inclusion area before updating under the current iteration number have an adjacent relation or not respectively, wherein when the current iteration number is the initial iteration number, the transition vertex set before updating under the current iteration number is the initial transition vertex set, and the inclusion area before updating under the current iteration number is the initial inclusion area; Each transition vertex with the adjacency relationship is incorporated into an incorporation region before updating under the current iteration number with the adjacency relationship, so that an incorporation region after updating under the current iteration number is obtained; Deleting all transition vertexes with the adjacency relation from the transition vertex set before updating under the current iteration number, and keeping all transition vertexes without the adjacency relation in the transition vertex set before updating under the current iteration number to obtain a transition vertex set after updating under the current iteration number; judging whether the number of the updated transition vertexes in the transition vertex set under the current iteration times is 0 or not; If not, determining the updated transition vertex set under the current iteration number as the updated transition vertex set under the next iteration number, determining the updated inclusion region under the current iteration number as the updated inclusion region under the next iteration number, updating the current iteration number as the next iteration number, and returning to judge whether each transition vertex in the updated transition vertex set under the current iteration number has an adjacent relation with any vertex in the updated inclusion region under the current iteration number; if so, determining the updated sulcus seed inclusion areas and the updated gyrus seed inclusion areas under the current iteration times as a sulcus seed target area and a gyrus seed target area respectively, thereby obtaining a brain personalized partition map and realizing the partition of the brain.
2. The method of individualizing a segmentation based on geometric features of a brain surface according to claim 1, wherein determining shape indices of vertices based on the three-dimensional mesh data comprises: Determining bending parameters of each vertex according to the three-dimensional grid data, wherein the bending parameters comprise Gaussian curvature and average curvature; determining the maximum principal vector and the minimum principal vector of each vertex according to the bending parameters of each vertex; and determining the shape index of each vertex according to the maximum principal vector and the minimum principal vector of each vertex.
3. The method of individualizing a segmentation based on geometric features of a brain surface according to claim 2, wherein determining bending parameters of each vertex from the three-dimensional mesh data comprises: Determining any vertex as a target vertex; calculating a bending parameter of a target vertex according to the three-dimensional grid data by using a bending parameter calculation formula, wherein the bending parameter calculation formula comprises: ; ; ; Wherein, the Gaussian curvature for the target vertex; forming the total area of the area for the target vertex and all adjacent vertices; the number of adjacent vertexes which are target vertexes; the vertex angle of a triangle formed by the target vertex, the ith adjacent vertex and the (i+1) th adjacent vertex is the (1) th adjacent vertex when i=N; the length of the edge formed by the target vertex and the ith adjacent vertex; The length of the edge formed by the target vertex and the (i+1) th adjacent vertex; The length of the edge formed by the ith adjacent vertex and the (i+1) th adjacent vertex of the target vertex; Is the average curvature of the target vertex; is a dihedral angle formed by the target vertex, the ith adjacent vertex and the (i+1) th adjacent vertex.
4. A method of individualizing a brain surface geometry according to claim 3, wherein determining a maximum principal vector and a minimum principal vector for each vertex based on the bending parameters of each vertex, respectively, comprises: calculating a maximum principal vector and a minimum principal vector of each vertex according to bending parameters of each vertex by using a principal vector calculation formula, wherein the principal vector calculation formula comprises: ; ; Wherein, the Is the maximum principal vector; is the minimum principal vector.
5. The method of individualizing a partition based on geometric features of a brain surface according to claim 4, wherein determining the shape index of each vertex based on the maximum principal vector and the minimum principal vector of each vertex, respectively, comprises: Calculating the shape index of each vertex according to the maximum principal vector and the minimum principal vector of each vertex by using a shape index calculation formula, wherein the shape index calculation formula comprises the following components: ; Wherein, the Is a shape index.
6. The method of personalized partitioning based on brain surface geometry of claim 1, wherein dividing each vertex into a sulcus seed point, a gyrus seed point, and remaining vertices based on shape indices of each vertex comprises: Determining any vertex as a current vertex; When the shape index of the current vertex is smaller than-0.5, determining the current vertex as a sulcus seed point; when the shape index of the current vertex is larger than 0.5, determining the current vertex as a brain return seed point; When the shape index of the current vertex is between-0.5 and 0.5, the current vertex is determined to be the remaining vertex.
7. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the brain surface geometry-based personalized partitioning method of any one of claims 1-6.
8. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the individualizing segmentation method based on geometrical features of the brain surface as claimed in any one of claims 1-6.
9. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the individualizing segmentation method based on geometrical features of the brain surface as claimed in any one of claims 1-6.

Description

Individuation partitioning method, device, medium and product based on brain surface geometric characteristics Technical Field The application relates to the technical field of neuroimaging, in particular to a brain surface geometric feature-based individuation partitioning method, device, medium and product. Background Currently, brain differentiation is mainly dependent on traditional brain patterns, such as Brodmann partitioning, etc. Although widely used, the standardized atlas has the following limitations that 1) the anatomical difference among individuals is ignored, individual characteristics cannot be reflected, 2) the standardized atlas is divided mainly based on two-dimensional slices, three-dimensional geometric information of brain surfaces is not fully utilized, 3) the subjectivity of manual division is strong, objective mathematical basis is lacked, and 4) the requirement of accurate medical treatment on individual analysis is difficult to adapt. Thus, the conventional brain segmentation method has low accuracy for brain segmentation. Therefore, developing an objective brain surface geometric feature-based individuation partitioning method has important scientific significance and application value. Disclosure of Invention The application aims to provide a brain surface geometric feature-based individuation partitioning method, device, medium and product, which are used for solving the problem of low accuracy of brain partitioning. In order to achieve the above object, the present application provides the following solutions: In a first aspect, the present application provides a method for individualizing a region based on geometric features of a brain surface, comprising: acquiring three-dimensional grid data of the brain surface, wherein the three-dimensional grid data comprises a plurality of vertexes and a plurality of edges connecting two adjacent vertexes; Determining a shape index of each vertex based on the three-dimensional mesh data; Dividing each vertex into a sulcus seed point, a brain return seed point and the rest vertex based on the shape index of each vertex; determining a plurality of initial areas of the sulcus seed based on the plurality of neighboring sulcus seed points, and determining a plurality of initial areas of the gyrus seed based on the plurality of neighboring gyrus seed points; determining connectivity between initial areas of the brain ditch seeds and connectivity between initial areas of the brain return seeds respectively, wherein the connectivity is connected or disconnected; Combining all the initial areas of the brain ditch seeds with connectivity as communication to obtain a plurality of brain ditch seed combining areas, and combining all the initial areas of the brain return seeds with connectivity as communication to obtain a plurality of brain return seed combining areas; determining all initial areas of the cerebral sulcus seeds which are not communicated as independent areas of the cerebral sulcus seeds, and determining all initial areas of the cerebral gyrus seeds which are not communicated as independent areas of the cerebral gyrus seeds; Determining all vertexes in the sulcus seed merging region, the sulcus seed independent region and the sulcus seed independent region with the vertex number smaller than the preset number and all the rest vertexes as transition vertexes to obtain an initial transition vertex set; Based on each initial inclusion region, iterative inclusion is carried out on each transition vertex in the initial transition vertex set, so that a plurality of brain ditch seed target regions and brain return seed target regions are obtained, and a brain personalized partition map is obtained, and the partition of the brain is realized. Optionally, determining a shape index of each vertex based on the three-dimensional mesh data includes: Determining bending parameters of each vertex according to the three-dimensional grid data, wherein the bending parameters comprise Gaussian curvature and average curvature; determining the maximum principal vector and the minimum principal vector of each vertex according to the bending parameters of each vertex; and determining the shape index of each vertex according to the maximum principal vector and the minimum principal vector of each vertex. Optionally, determining a bending parameter of each vertex according to the three-dimensional grid data includes: Determining any vertex as a target vertex; calculating a bending parameter of a target vertex according to the three-dimensional grid data by using a bending parameter calculation formula, wherein the bending parameter calculation formula comprises: Wherein K is the Gaussian curvature of the target vertex, S is the total area of the region formed by the target vertex and all the adjacent vertices, N is the number of adjacent vertices of the target vertex, θ i is the vertex angle of a triangle formed by the target vertex, the i-th adjacent vertex