Individualized brain parcellation with integrated funcitonal and morphological information

Hongming Li, Yong Fan

Abstract: We propose a novel brain parcellation method to define individualized, groupwise consistent brain network nodes based on integrated functional and morphological information. Particularly, our method is built under a collaborative multi-view clustering framework in conjunction with graph-regularization techniques. Our method is able to integrate multimodality imaging data in the brain parcellation to comprehensively capture inter-subject variability in functional anatomy. Our method has been evaluated based on resting state functional MRI and structural MRI data obtained from the WU-Minn Human Connectome Project. The experimental results have demonstrated that our method could obtain groupwise consistent and subject-specific parcellation results with better functional and structural coherence.