The Reconstruction Algorithm Of Neural Fibers In The Brain White Matter Based On Diffusion Tensor

The human brain is the most complex system in nature. And the nerve system isone of the most complex systems in the human brain. The distributing of the nerves isanfractuous and the diameter of the neural fiber is very small. So it is very difficult toaccurately describe the neural fibers in the brain using traditional medical techniques.One of the research focuses in magnetic resonance (MR) diffusion imaging is toreconstruct neural fibers in the brain white matter with the diffusion tensor imaging(DTI) data. And it is the unique way to reconstruct neural fibers in vivo. It can be usedin neurological diagnosis. And during the process of surgery, it can be used in surgicalnavigation to protect specific fiber structures. And it can combine with some othertechniques to study brain better. For example, it can combine with functionalmagnetic resonance imaging (fMRI) in brain cognitive research. In addition, the atlasof neural fibers helps further the understanding of brain structures and functions. Sothe development of this research is of great significance to the brain science.However, there are many difficulties to reconstruct fibers because of the limitationsof imaging technique and hardware. The data acquired from MRI is discrete. And thetypical voxel size used in the acquisition of the MR images is larger than thecharacteristic diameter of the neural fiber. The data that is used to reconstruct thefibers is affected by partial volume effect. An effective reconstruction algorithm,which is referred to as Correlative Vector Weighted Tracking (CVWT) method, isproposed in this paper. It uses the principal vector and fractional anisotropy (FA) ofthe neighbouring correlative voxels to estimate the direction of the fiber tracking line.As a result, this method can reflect the distribution of the neural fibers in the brainwhite matter better than other methods.