The Dynamic Functional Parcellation Of Brain Region And Its Application Based On FMRI Data

Cognitive science and brain science research show that the brain is a complex system with a high degree of organizational stratification.Functional connectivity analysis based on functional magnetic resonance imaging(fMRI)could be applied on functional parcellation of the brain and is the basis for understanding the structure and function of brain.Previous research on the functional connectivity parcellation rarely considered the dynamics of functional connections and could not reflect the dynamic interaction between brain functional sub-regions.In this paper,the hippocampus in subcontaneous tissue and visual area in cerebral cortex are selected as the regions of interest to research the functional parcellation method based on dynamic functional connectivity.The main contents of this paper are as follows:A parcellation method based on the dynamic functional connectivity state centroids.We performed static/dynamic functional connectivity analysis on the hippocampus and obtained static parcellation and dynamic parcellation results based on the dynamic functional connectivity state centroids,which were used to identify Alzheimer's disease.The results showed that the dynamic functional connectivity exhibited asymmetry of the brain and possessed a better classification effect in distinguishing Alzheimer's disease.Moreover,the reproducibility of dynamic parcellation results is higher.These results contributed to the development of brain regions and deepened our understanding of the structure and function of the hippocampus.A parcellation method based on the dynamic functional connectivity degree.Based on the dynamic functional connectivity degree of the brain,we used group independent component analysis(g-ICA)to extract brain network components and parcellated the visual cortex.Then we applied clustering analysis on the time series of each component.The degree of connectivity divides the visual cortex.The results showed that we extracted brain network components that were physiologically significant and similar to previous studies.In addition,the anti-correlation state obtained by time series analysis was related to the previous research results.These results deepened our understanding of the coordination and discoordination between functional networks and provided some new insights into brain parcellation.