Vertical pillow beam sub-branch segmentation method and device

Abstract

The invention provides a vertical occipital beam sub-branch segmentation method and device, and the method comprises the steps: obtaining a magnetic resonance image, determining a whole brain fiber streamline and an anatomical brain region segmentation image based on the magnetic resonance image, determining a vertical occipital beam fiber streamline based on the whole brain fiber streamline, and segmenting the vertical occipital beam sub-branch segmentation image based on the vertical occipital beam fiber streamline. Determining a sub-branch division region of interest based on the anatomical brain region segmentation map; determining a voxel mask matrix corresponding to each sub-branch division region of interest and a first endpoint coordinate and a second endpoint coordinate of the vertical occipital beam fiber streamline, and determining a nearest neighbor region of interest of the vertical occipital beam fiber streamline based on each voxel mask matrix, the first endpoint coordinate and the second endpoint coordinate; and determining a sub-branch to which the vertical pillow bundle fiber streamline belongs based on the nearest neighbor region of interest of the vertical pillow bundle fiber streamline. According to the invention, finer sub-branch segmentation of the vertical occipital beam can be realized.

Description

Technical Field

[0001] The invention relates to the technical field of magnetic resonance imaging, and in particular to a method and a device for segmenting a vertical occipital fasciculus sub-branch. Background Art

[0002] Diffusion MRI technology uses pulse gradients to generate a non-uniform linear magnetic field, so that the diffusion of water molecules along the gradient direction causes a certain phase difference in the microscopic magnetization vector, and obtains the signal intensity related to the degree of water molecule diffusion, thereby measuring the diffusion intensity of water molecules in the gradient direction and obtaining a set of diffusion-weighted image data. After modeling, the diffusion-weighted image can be expressed as a tensor model or a directional distribution function. The long axis or peak direction of the model indicates the direction in which the voxel diffuses most strongly, that is, the tissue structure has less obstruction along this direction; therefore, it can depict the direction of living tissue structure in a non-invasive way, and plays an important role in brain research. The special myelin structure in white matter makes the water molecules therein show restricted anisotropic diffusion. The diffusion direction in white matter can be obtained by using diffusion MRI technology, and the directions indicated by each voxel are smoothly connected, that is, the white matter fiber streamlines are obtained; the whole brain fibers are divided into different clusters according to geometric shapes, anatomical characteristics or functional characteristics, and white matter fiber bundles can be obtained.

[0003] The vertical occipital fasciculus is a vertical fiber located in the occipital lobe. It plays an important role in the visual pathway. It connects various cortical areas with different functions, integrates visual information related to object recognition, shape perception, color recognition, pattern recognition, visual object positioning and motion perception, and is an indispensable fiber bundle for visual analysis. However, when visual analysis is performed based on the vertical occipital fasciculus, the vertical occipital fasciculus at the occipital lobe is only regarded as a whole to screen vertical fibers from the whole brain fibers, determine the termination cortex, and remove the arcuate fasciculus fibers, etc., and the vertical occipital fasciculus is not further divided into subbranches. Since the vertical occipital fasciculus connects multiple cortical areas with different functions, when part of the fiber bundle structure changes, the overall impact on the vertical occipital fasciculus may not be significant, and the information transmitted and processed by different areas of the vertical occipital fasciculus is inconsistent, so some functions related to vision may be abnormal or changed; when visual analysis is performed based on the vertical occipital fasciculus, it is impossible to perform a more detailed subbranch division of the vertical occipital fasciculus, which makes it impossible to analyze the abnormalities of some functions related to vision. Therefore, how to perform more detailed sub-branch segmentation of the vertical occipital fasciculus and then complete the abnormal analysis of some vision-related functions is a technical problem that needs to be solved urgently. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method and apparatus for segmenting a vertical occipital fasciculus sub-branch to eliminate or improve one or more defects in the prior art.

[0005] One aspect of the present invention provides a method for segmenting a vertical occipital fasciculus sub-branch, the method comprising: Acquire a magnetic resonance image, determine a whole-brain fiber streamline and an anatomical brain region segmentation map based on the magnetic resonance image, determine a vertical occipital fasciculus fiber streamline based on the whole-brain fiber streamline, and determine a sub-branch division region of interest based on the anatomical brain region segmentation map; Determine the voxel mask matrix corresponding to each sub-branch divided region of interest and the first endpoint coordinates and the second endpoint coordinates of the vertical occipital fiber streamline, and determine the nearest neighbor region of interest of the vertical occipital fiber streamline based on each voxel mask matrix, the first endpoint coordinates and the second endpoint coordinates; The sub-branch to which the vertical occipital fascicle fiber streamline belongs is determined based on the nearest neighbor region of interest of the vertical occipital fascicle fiber streamline.

[0006] In some embodiments of the present invention, determining the nearest neighbor region of interest of the vertical occipital fiber streamline based on each of the voxel mask matrices, the first endpoint coordinates, and the second endpoint coordinates includes: The nearest neighbor region of interest of the vertical occipital fiber streamline is determined by a K-nearest neighbor algorithm based on each of the voxel mask matrices, the first endpoint coordinates, and the second endpoint coordinates.

[0007] In some embodiments of the present invention, determining the nearest neighbor region of interest of the vertical occipital fiber streamline by a K-nearest neighbor algorithm based on each of the voxel mask matrices, the first endpoint coordinates, and the second endpoint coordinates includes: Determine the center point coordinates of the regions of interest divided by each sub-branch based on each voxel mask matrix; Calculate each first distance between the first endpoint coordinate and each of the sub-branch divided regions of interest based on the first endpoint coordinate and each of the center point coordinates; Calculate each second distance between the second endpoint coordinate and each of the sub-branch divided regions of interest based on the second endpoint coordinate and each of the center point coordinates; The nearest neighbor region of interest of the vertical occipital fiber streamline is determined based on each of the first distances and each of the second distances.

[0008] In some embodiments of the present invention, determining the nearest neighbor region of interest of the vertical occipital fiber streamline based on each of the first distances and each of the second distances comprises: Determine a first region of interest corresponding to the first minimum distance value and a second region of interest corresponding to the second minimum distance value; When the first region of interest and the second region of interest are divided into regions of interest for only one sub-branch respectively, the first region of interest and the second region of interest are the nearest neighbor regions of interest of the vertical occipital fiber streamline.

[0009] In some embodiments of the present invention, the method further comprises: When the first region of interest and / or the second region of interest includes at least two regions of interest with the same distance, each third distance between the endpoint coordinates and each voxel point of each region of interest with the same distance is calculated based on the corresponding endpoint coordinates and the voxel mask matrix corresponding to the region of interest with the same distance, the third distance average value corresponding to each region of interest with the same distance is determined based on each of the third distances, and the final region of interest corresponding to the endpoint coordinates is determined based on the minimum third distance average value, and the final region of interest corresponding to the endpoint coordinates is the nearest neighbor region of interest of the vertical occipital fiber streamline.

[0010] In some embodiments of the present invention, determining the sub-branch to which the vertical occipital fascicle fiber streamline belongs based on the nearest neighbor region of interest of the vertical occipital fascicle fiber streamline comprises: It is determined whether the first region of interest and the second region of interest belong to the same functional sub-branch. When they belong to the same functional sub-branch, the functional sub-branch corresponding to the first region of interest and the second region of interest is the sub-branch to which the vertical occipital fiber streamline belongs.

[0011] In some embodiments of the present invention, the sub-branch division region of interest includes: ventral and dorsal regions of the first visual cortex V1, ventral and dorsal regions of the second visual cortex V2, ventral and dorsal regions of the third visual cortex V3, the inferior occipital sulcus region, the inferior occipital gyrus region, the transverse occipital sulcus region, the posterior transverse collateral sulcus region, the anterior occipital sulcus region, the collateral sulcus region, the posterior occipitotemporal sulcus region, and the posterior parietal sulcus region; The sub-branch is an early visual information encoding sub-bundle, a spatial feature selection sub-bundle, a motion information processing sub-bundle or a spatial attention sub-bundle; The sub-branch division areas of interest of the early visual information encoding sub-bundle include the ventral and dorsal areas of the first visual cortex V1, the ventral and dorsal areas of the second visual cortex V2, and the ventral and dorsal areas of the third visual cortex V3; the sub-branch division areas of interest of the spatial feature selection sub-bundle include the inferior occipital sulcus area, the inferior occipital gyrus area, and the transverse occipital sulcus area; the sub-branch division areas of interest of the motion information processing sub-bundle include the transverse occipital sulcus area, the posterior collateral transverse sulcus area, and the anterior occipital sulcus area; the sub-branch division areas of interest of the spatial attention sub-bundle include the collateral sulcus area, the posterior occipitotemporal sulcus area, and the posterior parietal sulcus area.

[0012] In some embodiments of the present invention, the method further comprises: Based on all the vertical occipital fiber streamlines in each subbranch, the vertical occipital fiber center streamlines corresponding to each subbranch are determined; Calculate the distances between each vertical occipital fiber streamline in each subbranch and the corresponding vertical occipital fiber center streamline; Determine the outlier fibers in each subbranch based on the distances between each vertical occipital fiber streamline in each subbranch and the corresponding vertical occipital fiber center streamline; The stray fibers are deleted from the corresponding sub-branch.

[0013] According to another aspect of the present invention, a vertical occipital fasciculus sub-branch segmentation device is also disclosed, the device comprising a processor, a memory and a computer program stored in the memory, the processor is used to execute the computer program, and when the computer program is executed, the device implements the steps of the method described in any of the above embodiments.

[0014] According to yet another aspect of the present invention, a computer-readable storage medium is disclosed, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described in any of the above embodiments are implemented.

[0015] The vertical occipital fascicle sub-branch segmentation method and device disclosed in the above-mentioned embodiment of the present application further determine the nearest region of interest of the vertical occipital fascicle fiber streamline based on the voxel mask matrix of the region of interest divided by each sub-branch, the first endpoint coordinates and the second endpoint coordinates of the vertical occipital fascicle fiber streamline, and determine the sub-branch to which the vertical occipital fascicle fiber streamline belongs according to the nearest region of interest of the vertical occipital fascicle fiber streamline; this method divides multiple vertical occipital fascicle fiber streamlines in the occipital lobe into sub-bundles with different functions in combination with the functional characteristics of the brain area connected to the vertical occipital fascicle, thereby achieving a more detailed sub-branch segmentation of the vertical occipital fascicle, and further completing the abnormal analysis of some functions related to vision.

[0016] Additional advantages, purposes, and features of the present invention will be described in part in the following description, and will become apparent to those skilled in the art after studying the following, or may be learned from the practice of the present invention. The purposes and other advantages of the present invention may be achieved and obtained by the structures specifically indicated in the specification and the accompanying drawings.

[0017] Those skilled in the art will appreciate that the objectives and advantages that can be achieved with the present invention are not limited to the above specific description, and the above and other objectives that can be achieved by the present invention will be more clearly understood from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of this application, and do not constitute a limitation of the present invention. The components in the drawings are not drawn to scale, but are only for illustrating the principles of the present invention. In order to facilitate the illustration and description of some parts of the present invention, the corresponding parts in the drawings may be enlarged, that is, they may become larger relative to other components in the exemplary device actually manufactured according to the present invention. In the drawings: Figure 1Schematic diagram of the process of a vertical occipital fasciculus sub-branch segmentation method according to an embodiment of the present application.

[0019] Figure 2 Schematic diagram of the process of a vertical occipital fasciculus sub-branch segmentation method according to another embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0021] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0022] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0023] It should also be noted here that, unless otherwise specified, the term "connection" in this article can refer not only to a direct connection, but also to an indirect connection with an intermediary, not only to a wired connection, but also to a wireless connection, and the specific connection can be changed based on the actual application scenario.

[0024] The following are the definitions of terms related to the technical solution: K-nearest neighbor algorithm: It is an instance-based supervised learning algorithm, commonly used to solve classification and regression problems. The algorithm finds the K known samples closest to it in the feature space, and determines the category or output value of the sample based on the categories or values ​​of these samples.

[0025] Vertical occipital fasciculus: It is an important fiber bundle in the visual pathway and is closely related to visual function. Retinal action potentials reach the lateral geniculate body of the thalamus through the optic nerve, optic chiasm, and optic tract, and then reach the primary visual cortex of the occipital lobe through the optic radiation. Visual information is initially encoded by cognitive processing within the primary visual cortex, and then divided into ventral and dorsal streams, reaching the temporal lobe and parietal lobe respectively; the ventral stream provides information about shape, color, and size to help identify and select targets, while the dorsal stream mediates information about spatial perception and goal-directed actions; the vertical occipital fasciculus is the main connecting bundle between these two branches, facilitating their close cooperation in various tasks.

[0026] Fiber tracking: The process of reconstructing brain white matter fibers from diffusion magnetic resonance images. This process is based on the characteristics of diffusion magnetic resonance imaging, which can detect the degree and direction of anisotropy of water molecules diffusing in tissues in vivo, estimate the distribution of fiber orientations in each voxel in the image, and further select the most likely orientation based on the probability of connection between the selected seed voxel and other voxels, thereby obtaining whole-brain fiber connections.

[0027] Anatomical brain atlas: A brain partition map used to analyze structural T1-weighted magnetic resonance images. Brain regions are often divided according to structure, function, etc. Different brain maps have different division rules; among the commonly used brain maps, the Desikan-Killiany map contains 68 cortical partitions, which retains the anatomical positions of the macroscopic sulci and gyri; the 2009 version of the Destrieux brain region template contains 74 cortical partitions in each of the left and right brains (a total of 148 partitions), providing a more detailed division for cortical research.

[0028] At present, when analyzing visual function based on magnetic resonance images, the prior art only segments the vertical occipital fascicle from the fiber streamlines of the whole brain according to the position and shape of the fiber bundles, that is, all the vertical occipital fascicles at the occipital lobe are regarded as a whole, and no more detailed sub-branch segmentation of the vertical occipital fascicle is performed based on the sub-functional characteristics of the visual function, thereby ignoring the abnormalities and changes brought by the fiber streamlines of each vertical occipital fascicle to the sub-functional characteristics of the visual function, thereby affecting the accuracy of the visual analysis results. In view of the above problems existing in the prior art, the present application proposes a vertical occipital fascicle sub-branch segmentation method and device, so as to further divide the region of interest of multiple vertical occipital fascicle fiber streamlines into multiple sub-bundles with different functions based on the sub-branches connected thereto, thereby achieving more detailed sub-branch segmentation of the vertical occipital fascicle.

[0029] In healthy adults, the ventral terminals of the vertical occipital fasciculus are concentrated in the inferior occipital gyrus (IOG), inferior occipital sulcus (IOS), and posterior transverse collateral sulcus (ptCoS). In addition, part of the fiber bundle extends to the midpoint of the mid-fusiform sulcus (MFS) and anterior occipital sulcus (aOS). The dorsal side of the vertical occipital fasciculus mainly falls in the transverse occipital sulcus (TOS) and posterior intraparietal sulcus (pIPS). In the ventral cortex to which it is connected, the inferior occipital gyrus and inferior occipital sulcus are located upstream of the ventral pathway, have spatial and facial feature selection functions, and provide the neural basis for the interaction between features and positions in perception; the posterior collateral sulcus is part of the fourth visual cortex V4, usually as an intermediate stage of pattern recognition, and is related to shape processing; the anterior occipital sulcus, as a temporal visual area, is responsible for extracting depth information from motion perception and performing multimodal motion processing. In addition, the V4 area is also related to motion processing, and it coordinates with the dorsal related areas to jointly affect motion perception. The middle fusiform sulcus is composed of the collateral sulcus (CoS) and the posterior occipitotemporal sulcus (pOTS); the collateral sulcus is located at the lateral edge of the parahippocampal gyrus, where the object recognition visual information from the occipitotemporal lobe and the spatial context information from the parietal lobe intersect, and is the intersection area of ​​the dorsal stream and the ventral stream, responsible for controlling the complex mechanism of spatial memory; the posterior occipitotemporal sulcus is usually called the visual word form area, which plays an important role in the cognitive process of shape processing.

[0030] In the dorsal cortex, the area adjacent to the medial side of the transverse occipital sulcus is a subarea of ​​the third visual cortex, V3A, which is related to motion processing and spatial attention and is important in scene processing. In addition, the posterior parietal sulcus also has multiple functions, including the management of spatial attention, the coordination of eye movements, and the promotion of intrinsic and exogenous orientation processes.

[0031] In order to facilitate a better description of the technical solution of the present application, based on the functions of the above-mentioned brain regions, the present application subdivides the vertical occipital fasciculus into four functional sub-fascicules (sub-branches), namely: (1) vertical occipital fasciculus I (also known as the early visual information encoding sub-fasciculus): responsible for the encoding of early visual information, including vertical fibers connecting the ventral and dorsal areas of the first visual cortex V1 to the third visual cortex V3; (2) vertical occipital fasciculus II (also known as the spatial feature selection sub-fasciculus): responsible for the selection of spatial feature information, including vertical fibers connecting the inferior occipital sulcus, inferior occipital gyrus and transverse occipital sulcus; (3) vertical occipital fasciculus III (also known as the motion information processing sub-fasciculus): responsible for the processing of motion-related information, including vertical fibers connecting the posterior collateral sulcus, anterior occipital sulcus and transverse occipital sulcus; (4) vertical occipital fasciculus IV (also known as the spatial attention sub-fasciculus): spatial attention, including vertical fibers connecting the middle fusiform sulcus (collateral sulcus and posterior occipitotemporal sulcus) and the posterior parietal sulcus. However, it can be understood that the number of functional sub-beams listed above and the functions corresponding to each functional sub-beam are only examples. In other embodiments, the vertical pillow beam can be further subdivided into other numbers of functional sub-beams, and the functions corresponding to each functional sub-beam can also be other.

[0032] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0033] Figure 1 FIG. 1 is a flow chart of a vertical occipital fasciculus sub-branch segmentation method according to an embodiment of the present application. Figure 1 As shown, the vertical occipital fasciculus sub-branch segmentation method at least includes steps S10 to S30.

[0034] Step S10: Acquire a magnetic resonance image, determine the whole-brain fiber streamlines and the anatomical brain region segmentation map based on the magnetic resonance image, determine the vertical occipital fasciculus fiber streamlines based on the whole-brain fiber streamlines, and determine the sub-branch division region of interest based on the anatomical brain region segmentation map.

[0035] This step is to trace the whole-brain fiber streamlines from the diffusion magnetic resonance image, and segment the vertical occipital fasciculus fiber streamlines located in the occipital lobe from the whole-brain fiber streamlines, providing candidate fibers for subsequent sub-branch segmentation; in addition, the anatomical brain regions are segmented according to the magnetic resonance image to obtain an anatomical brain region segmentation map, providing sub-branch division of interest regions for sub-branch segmentation; exemplarily, the anatomical brain regions can be segmented based on the T1 structural magnetic resonance image.

[0036] When performing whole-brain fiber tracking on magnetic resonance images, for example, a multi-shell multi-tissue constrained spherical deconvolution algorithm can be used to estimate the white matter fiber direction distribution function from diffusion magnetic resonance images in different directions, place seed points in white matter voxels, and use the probabilistic fiber tracking technology in the MRtrix3 neuroimaging visualization tool to generate whole-brain fiber streamlines.

[0037] For anatomical brain segmentation, the T1 structural MRI images were registered to Talairach space (also known as Taylor-Rush space or Tylorach coordinate system) by manually marking the anterior commissure (AC) and posterior commissure (PC). The T1 images were automatically segmented based on the Destrieux brain atlas using the Freesurfer tool.

[0038] When segmenting the vertical occipital fasciculus from the whole-brain fiber streamline, the fiber automatic quantitative AFQ package of matlab software can be used to read the coordinates of the whole-brain fiber streamline, and the arcuate fasciculus can be identified using the traditional automated fiber quantitative analysis method (AFQ method). The coordinates of the ventral occipital temporal cortex (including the fusiform gyrus, inferior occipital gyrus, and inferior temporal gyrus) in the anatomical brain region segmentation are read, and all fibers terminating within 2mm of the ventral occipital temporal cortex are taken as candidate fibers. Furthermore, candidate fibers with vertical extension directions are retained, and fibers mixed with the arcuate fasciculus are removed; the linear bundle estimation algorithm LiFE is used to evaluate the degree of matching of each fiber relative to the diffusion direction of the corresponding voxel, and fibers with a probability less than 0 are removed to optimize the segmentation results of the vertical occipital fasciculus.

[0039] When configuring the environment, development can be carried out based on the matlab programming language, and data processing and whole-brain fiber tracking rely on the ubuntu18.04 system; among them, the Freesurfer tool can be configured in the ubuntu system for anatomical brain area segmentation, and FSL and MRtrix3 can be configured for diffusion magnetic resonance data processing and whole-brain fiber tracking; the automatic fiber quantification package AFQ can be configured in the matlab software for vertical occipital fasciculus tracking.

[0040] In some embodiments of the present invention, the brain regions required for sub-branch segmentation are matched from the anatomical brain region segmentation map as the sub-branch segmentation regions of interest, and the sub-branch segmentation regions of interest may include: the ventral and dorsal regions of the first visual cortex V1, the ventral and dorsal regions of the second visual cortex V2, the ventral and dorsal regions of the third visual cortex V3, the inferior occipital sulcus region, the inferior occipital gyrus region, the transverse occipital sulcus region, the posterior transverse collateral sulcus region, the anterior occipital sulcus region, the collateral sulcus region, the posterior occipital-temporal sulcus region, and the posterior parietal sulcus region. The ventral and dorsal regions of the visual cortex V1-V3 are determined by Mona's human occipito-temporal visual cortex atlas, including 6 regions from the occipital pole to the transverse occipital sulcus, and there are 12 regions in the left and right brains; and the other regions of interest are selected from the Destrieux brain atlas according to their locations and names. If there are two or more regions of interest in the same brain region, the brain region will be used as the sub-branch segmentation region of interest, such as the inferior occipital sulcus and inferior occipital gyrus in the atlas are the same brain region, and the brain region number in Freesurfer is 11102; if the naming of the brain atlas is inconsistent with the region of interest, the corresponding brain region is selected in the brain atlas according to the location of the region of interest. The brain region numbers corresponding to the regions of interest are shown in the following table. Step S20: Determine the voxel mask matrix corresponding to each of the sub-branch divided regions of interest and the first endpoint coordinates and the second endpoint coordinates of the vertical occipital fiber streamline, and determine the nearest neighbor region of interest of the vertical occipital fiber streamline based on each of the voxel mask matrix, the first endpoint coordinates and the second endpoint coordinates.

[0041] In this step, the regions of interest for each sub-branch division are converted from surface mapping into voxel masks, and the ventral and dorsal visual areas of the Mona atlas are converted from the standard space to the individual space of the subject; the coordinate data of the vertical occipital fiber streamlines are determined, and it is ensured that the vertical occipital fiber streamlines and the sub-branch division regions of interest are located in the same space.

[0042] The endpoint coordinates of all vertical occipital fiber streamlines are traversed, and the nearest neighbor region of interest of each vertical occipital fiber streamline is determined based on the voxel mask matrix corresponding to the region of interest divided by each sub-branch and the first endpoint coordinates and the second endpoint coordinates of each vertical occipital fiber streamline.

[0043] Exemplarily, determining the nearest neighbor region of interest of the vertical occipital fiber streamline based on each of the voxel mask matrices, the first endpoint coordinates, and the second endpoint coordinates includes: determining the nearest neighbor region of interest of the vertical occipital fiber streamline through the K nearest neighbor algorithm based on each of the voxel mask matrices, the first endpoint coordinates, and the second endpoint coordinates. Specifically, determining the nearest neighbor region of interest of the vertical occipital fiber streamline through the K nearest neighbor algorithm based on each of the voxel mask matrices, the first endpoint coordinates, and the second endpoint coordinates specifically includes: determining the center point coordinates of each sub-branch divided region of interest based on each of the voxel mask matrices; calculating each first distance between the first endpoint coordinate and each sub-branch divided region of interest based on the first endpoint coordinates and each of the center point coordinates; calculating each second distance between the second endpoint coordinate and each of the sub-branch divided region of interest based on the second endpoint coordinates and each of the center point coordinates; determining the nearest neighbor region of interest of the vertical occipital fiber streamline based on each of the first distances and each of the second distances. In this embodiment, the distances between each endpoint of the vertical occipital fiber streamline and each sub-branch divided region of interest are calculated, so that the sub-branch divided region of interest with the closest distance to the vertical occipital fiber streamline is taken as its nearest neighbor region of interest.

[0044] Further, determining the nearest neighbor region of interest of the vertical occipital fiber streamline based on each of the first distances and each of the second distances also includes the following steps: determining a first region of interest corresponding to the minimum value of the first distance and a second region of interest corresponding to the minimum value of the second distance; when the first region of interest and the second region of interest are each divided into regions of interest by only one sub-branch, the first region of interest and the second region of interest are the nearest neighbor regions of interest of the vertical occipital fiber streamline. In this embodiment, the first region of interest and the second region of interest each have only one sub-branch divided region of interest, and at this time, the corresponding sub-branch divided region of interest is directly used as the nearest neighbor region of interest of the vertical occipital fiber streamline.

[0045] In addition, if the first region of interest and / or the second region of interest includes at least two regions of interest with the same distance, each third distance between the endpoint coordinates and each voxel point of each region of interest with the same distance is calculated based on the corresponding endpoint coordinates and the voxel mask matrix corresponding to the region of interest with the same distance, and the third distance average value corresponding to each region of interest with the same distance is determined based on each of the third distances, and the final region of interest corresponding to the endpoint coordinate is determined based on the minimum third distance average value, and the final region of interest corresponding to the endpoint coordinate is the nearest neighbor region of interest of the vertical occipital fiber streamline. In this embodiment, the distances between the multiple sub-branch divided regions of interest and the endpoints of the vertical occipital fiber streamlines are equal. At this time, the distance calculation is further performed based on the coordinates of other points (such as the coordinates of the edge points of the region) other than the center point coordinates of the sub-branch divided regions of interest and the endpoint coordinates of the vertical occipital fiber streamlines; for example, when the first region of interest includes two regions of interest with the same distance, the third distance between the coordinates of the first endpoint and each region of interest with the same distance is further calculated. When calculating the third distance, the coordinates of the sub-branch divided regions of interest used are other coordinates except the center point coordinates, such as the edge point coordinates, etc. At this time, the third distance corresponding to each region of interest with the same distance is a plurality of values. In order to determine the final distance between each region of interest with the same distance and the first endpoint, the plurality of third distance values ​​are averaged, and then the region of interest with the same distance corresponding to the minimum third distance average value is used as the final region of interest of the first endpoint.

[0046] In a specific embodiment, a K-dimensional search tree can be constructed based on the coordinates of the regions of interest divided by each sub-branch, so as to search the regions of interest closest to the two ends of each vertical occipital fiber streamline in the K-dimensional tree searcher to separate the fibers. That is, the reading of the voxel mask matrix corresponding to the regions of interest divided by each sub-branch is realized based on the matlab code, and each region of interest divided by each sub-branch represents a category, and the initial input data set is constructed. In order to efficiently perform the nearest neighbor search, the built-in function KDTreeSearcher of matlab is used to organize the data into a K-dimensional tree searcher suitable for the nearest neighbor search; the K-dimensional tree is a binary tree structure that partitions the data space, and the data is divided into different subspaces by recursively selecting dimensions and partitioning points; the K-dimensional tree searcher uses the hierarchical structure of the tree to quickly narrow the search range and only access potential nearest neighbor candidate regions, thereby improving the search efficiency. When searching for the nearest neighbor region of interest of the vertical occipital fiber streamline, the coordinate points of the vertical occipital fiber streamline can be read, and the endpoint coordinates of all vertical occipital fiber streamlines can be traversed, and the nearest neighbor search of the K-dimensional tree can be performed through the knnsearch function. Specifically, the Euclidean distances between the endpoint coordinates of all sub-branch division regions of interest and the vertical occipital fiber streamlines are calculated, and the sub-branch division region of interest closest to the vertical occipital fiber streamlines is found based on the calculated distance values. The distance calculation formula is: , D represents the Euclidean distance between two points, is the coordinate value of the endpoint of the vertical pillow bundle fiber streamline i in three-dimensional space, represents the coordinates of the points where the t-th subbranch divides the region of interest. Based on the above, we can know that It can represent the coordinates of the center point or other points of the sub-branch division area of ​​interest (such as the coordinates of the edge point of the area).

[0047] In addition, after determining the nearest neighbor region of interest of each endpoint, it is further judged whether the distance between the endpoint and the nearest neighbor region of interest exceeds 4 cm. When the distance exceeds 4 cm, it is considered that the vertical occipital fiber streamline does not belong to any sub-branch, and the vertical occipital fiber streamline is deleted from the candidate fibers.

[0048] Obviously, the above embodiments using the K nearest neighbor algorithm to determine the nearest neighbor region of interest of the vertical occipital fiber streamline are only some examples. In some other embodiments, other algorithms can also be used to determine the nearest neighbor region of interest of the vertical occipital fiber streamline.

[0049] Step S30: determining the sub-branch to which the vertical occipital fiber streamline belongs based on the nearest neighbor region of interest of the vertical occipital fiber streamline.

[0050] In this step, the sub-branch to which the vertical occipital fiber streamline belongs is further determined based on the nearest neighbor region of interest of the two end points of the vertical occipital fiber streamline. In some embodiments, the sub-branch can be divided based on the function of the brain region, for example, the ventral and dorsal brain regions are divided into four groups according to the function of the brain region, and the vertical occipital fiber streamlines connecting the brain regions in the group are divided into corresponding functional groups to obtain functional sub-bundles of each sub-branch; in this embodiment, the fibers connecting the brain regions in the functional sub-bundles are the pathways in the vertical occipital fascicles that transmit information related to the function. Specifically, the ends of each vertical occipital fiber streamline will be divided into two ventral and dorsal brain regions, and only when the ventral and dorsal brain regions of the fiber streamline belong to the same functional sub-bundle (or sub-branch), the fiber streamline will be added to the corresponding sub-bundle. Exemplarily, the four sub-branches are: early visual information encoding sub-bundle, spatial feature selection sub-bundle, motion information processing sub-bundle, and spatial attention sub-bundle.

[0051] The following table is a comparison of the brain areas and functions connected to the vertical occipital fasciculus sub-branches: In one embodiment, determining the sub-branch to which the vertical occipital fiber streamline belongs based on the nearest neighbor region of interest of the vertical occipital fiber streamline includes: judging whether the first region of interest and the second region of interest belong to the same functional sub-branch, and when they belong to the same functional sub-branch, the functional sub-branch corresponding to the first region of interest and the second region of interest is the sub-branch to which the vertical occipital fiber streamline belongs. For example, in the above table, the sub-branch division region of the early visual information encoding sub-bundle includes the ventral and dorsal regions of the first visual cortex V1, the ventral and dorsal regions of the second visual cortex V2, and the ventral and dorsal regions of the third visual cortex V3, the sub-branch division region of the spatial feature selection sub-bundle includes the inferior occipital sulcus region, the inferior occipital gyrus region, and the transverse occipital sulcus region, the sub-branch division region of the motion information processing sub-bundle includes the transverse occipital sulcus region, the posterior transverse collateral sulcus region, and the anterior occipital sulcus region, and the sub-branch division region of the spatial attention sub-bundle includes the collateral sulcus region, the posterior occipital-temporal sulcus region, and the posterior parietal sulcus region. Exemplarily, if the nearest neighbor regions of interest of the two endpoints of a vertical occipital fascicle fiber streamline are the inferior occipital gyrus region and the transverse occipital sulcus region, respectively, then the vertical occipital fascicle fiber streamline is divided into the spatial feature selection sub-bundle; if the nearest neighbor regions of interest of the two endpoints of a vertical occipital fascicle fiber streamline are the inferior occipital gyrus region and the posterior parietal sulcus region, respectively, at this time, since the inferior occipital gyrus region and the posterior parietal sulcus region do not belong to the same functional sub-branch, the vertical occipital fascicle fiber streamline does not belong to any sub-branch and can be directly deleted from the vertical occipital fascicle to be divided.

[0052] In some embodiments of the present invention, the method further includes: determining the vertical occipital fiber center streamline corresponding to each sub-branch based on all vertical occipital fiber streamlines in each sub-branch; calculating the distances between each vertical occipital fiber streamline in each sub-branch and the corresponding vertical occipital fiber center streamline; determining the outlier fiber in each sub-branch based on the distances between each vertical occipital fiber streamline in each sub-branch and the corresponding vertical occipital fiber center streamline; and deleting the outlier fiber from the corresponding sub-branch. Figure 2 As shown in the figure, after the initial segmentation of the sub-branch, the structure of some sub-bundles is not regular and there are some outlier fibers. In order to optimize the structure of the sub-branch, the fiber streamlines that deviate from the center of the sub-bundle in each sub-branch are further deleted, thereby obtaining the optimized sub-branch segmentation result.

[0053] Specifically, firstly, linear interpolation resampling was performed on all vertical occipital fiber streamlines in the subbranch, and the number of nodes of each fiber streamline was unified to 100; the coordinates of the center point of each fiber streamline in the subbranch on the corresponding node were calculated to obtain the center line of the fiber bundle; the Euclidean distance between each node of each fiber streamline and the corresponding node of the center line was traversed and calculated, and outlier fibers whose average distance from the center line exceeded 2 mm were removed.

[0054] Through the above embodiments, it can be found that the vertical occipital fasciculus, as a large connecting fasciculus connecting multiple brain regions, has rich functional characteristics in the brain regions it connects, resulting in different functional sub-branches. Therefore, the present application starts with the function of the brain regions connected to the vertical occipital fasciculus, and uses the distance from the fiber streamlines to each brain region to divide it into sub-bundles with different functions, breaking through the limitation of the traditional method of segmenting only according to the position and shape of the fiber bundle, thereby achieving detailed sub-branch segmentation of the vertical occipital fasciculus. This division method based on functional characteristics is more in line with the structural characteristics of the actual brain function network and can support more in-depth brain function research. And based on this method, the vertical occipital fasciculus is segmented in detail, avoiding the analysis deviation caused by the possible neglect of local fiber abnormalities or changes, and improving the accuracy and reliability of subsequent visual analysis or brain function research.

[0055] Correspondingly, the present invention also provides a vertical occipital fasciculus sub-branch segmentation device, which includes a processor, a memory, and a computer program stored in the memory, wherein the processor is used to execute the computer program, and when the computer program is executed, the device implements the steps of the method described in any of the above embodiments.

[0056] The embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described in any of the above embodiments are implemented. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0057] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0058] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0059] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with features of other embodiments or replace features of other embodiments.

[0060] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A vertical occipital fasciculus sub-branch segmentation method, characterized in that: The method comprises: Acquire a magnetic resonance image, determine a whole-brain fiber streamline and an anatomical brain region segmentation map based on the magnetic resonance image, determine a vertical occipital fasciculus fiber streamline based on the whole-brain fiber streamline, and determine a sub-branch division region of interest based on the anatomical brain region segmentation map; Determine the voxel mask matrix corresponding to each sub-branch divided region of interest and the first endpoint coordinates and the second endpoint coordinates of the vertical occipital fiber streamline, and determine the nearest neighbor region of interest of the vertical occipital fiber streamline based on each voxel mask matrix, the first endpoint coordinates and the second endpoint coordinates; The sub-branch to which the vertical occipital fascicle fiber streamline belongs is determined based on the nearest neighbor region of interest of the vertical occipital fascicle fiber streamline.

2. The vertical occipital fasciculus sub-branch segmentation method according to claim 1, characterized in that: Determining the nearest neighbor region of interest of the vertical occipital fiber streamline based on each of the voxel mask matrices, the first endpoint coordinates, and the second endpoint coordinates, comprises: The nearest neighbor region of interest of the vertical occipital fiber streamline is determined by a K-nearest neighbor algorithm based on each of the voxel mask matrices, the first endpoint coordinates, and the second endpoint coordinates.

3. The vertical occipital fasciculus sub-branch segmentation method according to claim 2, characterized in that: Determining the nearest neighbor region of interest of the vertical occipital fiber streamline by a K-nearest neighbor algorithm based on each of the voxel mask matrices, the first endpoint coordinates, and the second endpoint coordinates includes: Determine the center point coordinates of the regions of interest divided by each sub-branch based on each voxel mask matrix; Calculate each first distance between the first endpoint coordinate and each of the sub-branch divided regions of interest based on the first endpoint coordinate and each of the center point coordinates; Calculate each second distance between the second endpoint coordinate and each of the sub-branch divided regions of interest based on the second endpoint coordinate and each of the center point coordinates; The nearest neighbor region of interest of the vertical occipital fiber streamline is determined based on each of the first distances and each of the second distances.

4. The vertical occipital fasciculus sub-branch segmentation method according to claim 3, characterized in that: Determining the nearest neighbor region of interest of the vertical occipital fiber streamline based on each of the first distances and each of the second distances includes: Determine a first region of interest corresponding to the first minimum distance value and a second region of interest corresponding to the second minimum distance value; When the first region of interest and the second region of interest are divided into regions of interest for only one sub-branch respectively, the first region of interest and the second region of interest are the nearest neighbor regions of interest of the vertical occipital fiber streamline.

5. The vertical occipital fasciculus sub-branch segmentation method according to claim 4, characterized in that: The method further comprises: When the first region of interest and / or the second region of interest includes at least two regions of interest with the same distance, each third distance between the endpoint coordinates and each voxel point of each region of interest with the same distance is calculated based on the corresponding endpoint coordinates and the voxel mask matrix corresponding to the region of interest with the same distance, the third distance average value corresponding to each region of interest with the same distance is determined based on each of the third distances, and the final region of interest corresponding to the endpoint coordinates is determined based on the minimum third distance average value, and the final region of interest corresponding to the endpoint coordinates is the nearest neighbor region of interest of the vertical occipital fiber streamline.

6. The vertical occipital fasciculus sub-branch segmentation method according to claim 4, characterized in that: Determining the sub-branch to which the vertical occipital fascicle fiber streamline belongs based on the nearest neighbor region of interest of the vertical occipital fascicle fiber streamline, comprising: It is determined whether the first region of interest and the second region of interest belong to the same functional sub-branch. When they belong to the same functional sub-branch, the functional sub-branch corresponding to the first region of interest and the second region of interest is the sub-branch to which the vertical occipital fiber streamline belongs.

7. The vertical occipital fasciculus sub-branch segmentation method according to any one of claims 1 to 6, characterized in that: The sub-branch division areas of interest include: ventral and dorsal areas of the first visual cortex V1, ventral and dorsal areas of the second visual cortex V2, ventral and dorsal areas of the third visual cortex V3, the inferior occipital sulcus area, the inferior occipital gyrus area, the transverse occipital sulcus area, the posterior transverse collateral sulcus area, the anterior occipital sulcus area, the collateral sulcus area, the posterior occipitotemporal sulcus area, and the posterior parietal sulcus area; The sub-branch is an early visual information encoding sub-bundle, a spatial feature selection sub-bundle, a motion information processing sub-bundle or a spatial attention sub-bundle; The sub-branch division areas of interest of the early visual information encoding sub-bundle include the ventral and dorsal areas of the first visual cortex V1, the ventral and dorsal areas of the second visual cortex V2, and the ventral and dorsal areas of the third visual cortex V3; the sub-branch division areas of interest of the spatial feature selection sub-bundle include the inferior occipital sulcus area, the inferior occipital gyrus area, and the transverse occipital sulcus area; the sub-branch division areas of interest of the motion information processing sub-bundle include the transverse occipital sulcus area, the posterior collateral transverse sulcus area, and the anterior occipital sulcus area; the sub-branch division areas of interest of the spatial attention sub-bundle include the collateral sulcus area, the posterior occipitotemporal sulcus area, and the posterior parietal sulcus area.

8. The vertical occipital fasciculus sub-branch segmentation method according to claim 7, characterized in that: The method further comprises: Based on all the vertical occipital fiber streamlines in each subbranch, the vertical occipital fiber center streamlines corresponding to each subbranch are determined; Calculate the distances between each vertical occipital fiber streamline in each subbranch and the corresponding vertical occipital fiber center streamline; Determine the outlier fibers in each subbranch based on the distances between each vertical occipital fiber streamline in each subbranch and the corresponding vertical occipital fiber center streamline; The stray fibers are deleted from the corresponding sub-branch.

9. A vertical occipital fasciculus subbranch segmentation device, the device comprising a processor, a memory, and a computer program stored in the memory, characterized in that: The processor is used to execute the computer program. When the computer program is executed, the device implements the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images