Method and apparatus of fiber tracking, and non-transitory computer-readable medium thereof

A non-transitory computer-readable medium for fiber tracking, configured to store operating instructions which are executed by at least one processor to execute following operations:

• receiving a first diffusion weighted image of a subject's brain;
• obtaining N tracking parameters from a first relation matrix stored in a storage unit according to first N of all similarities of the first relation matrix, wherein N is a positive integer greater than or equal to 8;
• performing a fiber tracking process on the first diffusion weighted image according to each of the N tracking parameters, to obtain a plurality of first tracking images;
• obtaining two first regions of interest according to a predetermined fasciculus;
• extracting a plurality of nerve fibers between the two first regions of interest in each of the plurality of first tracking images;
• according to the plurality of nerve fibers extracted from the plurality of first tracking images, obtaining nerve fasciculus probabilities of overlapped voxels based on an overlapping equation, wherein the overlapping equation is Tj=norm(Σi=1n wi×(tj)i), and Tj is a nerve fasciculus probability of the j-th overlapped voxel, wi is a similarity corresponding to the i-th predetermined tracking parameter, (tj)i is a quantity of the nerve fibers of the j-th image voxel of the first tracking image obtained according to the i-th predetermined tracking parameter, and the j-th overlapped voxel corresponds in position to the j-th image voxel of each of the plurality of first tracking images, and i and j are positive integers;
• among the overlapped voxels, selecting the overlapped voxels having the top one to ten percent of nerve fasciculus probabilities to establish a nerve fasciculus skeleton;
• and
• comparing structural similarities between the nerve fasciculus skeleton and the extracted nerve fibers in each of the plurality of first tracking images, to obtain a plurality of evaluation values, and then obtaining a specific fasciculus according to the nerve fibers having the top thirty-three to sixty-six percent of the evaluation values.

An apparatus of fiber tracking, comprising:

• at least one processor;
• a storage unit;
• and
• at least one program stored in the storage unit and configured to be executed by the at least one processor, and at least one program comprises instructions for operations of:
• receiving a first diffusion weighted image of a subject's brain;
• obtaining N tracking parameters from a first relation matrix stored in a storage unit according to first N of all similarities of the first relation matrix, wherein N is a positive integer greater than or equal to 8;
• performing a fiber tracking process on the first diffusion weighted image according to each of the N tracking parameters, to obtain a plurality of first tracking images;
• obtaining two first regions of interest according to a predetermined fasciculus;
• extracting a plurality of nerve fibers between the two first regions of interest in each of the plurality of first tracking images;
• according to the plurality of nerve fibers extracted from the plurality of first tracking images, obtaining nerve fasciculus probabilities of overlapped voxels based on an overlapping equation, wherein the overlapping equation is Tj=norm(Σi=1n wi×(tj)i), and Tj is a nerve fasciculus probability of the j-th overlapped voxel, wi is a similarity corresponding to the i-th predetermined tracking parameter, (tj)i is a quantity of the nerve fibers of the j-th image voxel of the first tracking image obtained according to the i-th predetermined tracking parameter, and the j-th overlapped voxel corresponds in position to the j-th image voxel of each of the plurality of first tracking images, and i and j are positive integers;
• among the overlapped voxels, selecting the overlapped voxels having the top one to ten percent of nerve fasciculus probabilities to establish a nerve fasciculus skeleton;
• and
• comparing structural similarities between the nerve fasciculus skeleton and the extracted nerve fibers in each of the plurality of first tracking images, to obtain a plurality of evaluation values, and then obtaining a specific fasciculus according to the nerve fibers having the top thirty-three to sixty-six percent of the evaluation values.