Brain Structural Network Construction Based On Diffusion Magnetic Resonance Imaging

Human brain is one of the most complex systems in nature,and brain science has always been a hot area for scientists to study.In recent years,the rapid development of magnetic resonance imaging technology,white matter fiber tracking technology and brain structure network construction not only makes people have a deeper understanding of the complex fiber bundle path in the brain and the network connection of each brain interval,but also Clinical medicine provides a strong theoretical basis and technical means for the diagnosis and prevention of various neurological diseases.This paper combines the National Natural Science Foundation of China and Zhejiang Province to study the construction of brain structure networks based on diffusion magnetic resonance imaging.Firstly,this paper briefly describes the research progress and current status of diffusion magnetic resonance imaging,nerve fiber bundle tracking technology and brain structure network construction at home and abroad,and introduces the basic imaging principles of magnetic resonance imaging and diffusion magnetic resonance imaging.DTI,q-ball imaging,and SPFI imaging were used to reconstruct the direction of the nerve fibers corresponding to the single voxels in brain regions,and the performance of the three reconstruction schemes were compared on Fiber Cup simulation data set and real brain data set.Secondly,this paper proposes a new fiber tracking algorithm based on non-uniform rational B-spline(NURBS)Curve Fitting to track white matter fiber paths.Based on the Fiber Cup dataset and HCP dataset,local point-to-point performance indicators and global connectivity indicators were used,and the tracking performance of the traditional fiber tracking algorithm FACT was compared in the three ROI regions of the corpus callosum,parietal lobe and temporal lobe.The results showed that the fiber path tracked by the NURBS algorithm presented in this paper showed good anatomical consistency.Thirdly,AAL brain region template was used to divide the cerebral cortex into 116 different brain regions.Each brain region represents a node in the network.The number of fibers tracked in each brain segment is used as the edge of the network to construct a brain structure network.The topological parameters of the brain structure network are calculated and analyzed.The results show that the network has high global efficiency,short minimum path length and "small world" attribute.Finally,this paper analyzes the topological characteristics of the brain network of Alzheimer's patients,and uses four brain region templates to construct the brain structure network of AD patients with different disease levels.The results show that the topological properties of brainstructure network change significantly with the severity of disease,and the HCP-MMP profile template can significantly show the difference between various topological parameters,which provides an important reference for the prevention and diagnosis of AD disease.