Connectivity-based Parcellation Of The Human Right Brodmann Area 46

With the development of brain science, the brain projects such as genome project, human connectome project and brainnetome project have emerged in recent years. The brain atlas is the core and basis of the human connectome project and brainnetome project. Because of intrinsic limitations of the existing atlas, mounting studies have many errors. With the advance in magnetic resonance imaging(MRI), to explore the functional organization of the brain based on multimodal neuroimaging technique is the focus of the brain research. Recently, the connectivity-based parcellation of specific brain area greatly promotes the study of brain atlas. Diffusion tensor imaging(DTI) can non-invasively characterize the white matter pathway of the brain providing the technique support for constructing brain atlas in vivo. Moreover, on the basis of newly constructed brain atlas, we can reveal the brain mechanism for specific cognitive function from a new perspective.Although most of the previous studies considered the BA46 area as a functionally homogenous brain region, investigation on this area suggests that BA46 area not only involved in working memory, but also in motion control, speech processing, mental arithmetic and other functions. In addition, the cytoarchitecture-based mapping of macaque BA46 identified different subregions in this area. these studies indicating that different subregions maybe exist in BA46 and each subregion has unique functional contribution. Thus, in this study, we aimed to identify functional subregions of BA46 in human brain and to establish its functional organization.In this thesis, we used connectivity-based parcellation approach and edge-weighted spectral clustering algorithm to subdivide the human BA46. The Pearson Cramer’s V coefficient was employed to determine the cluster numbers. Moreover, we used fiber tracking method and resting-state functional connectivity analyses to map the main fiber pathway and cortical network for each subregion. We additionally calculated the functional connectivity with specific targets to obtain the fingerprints to characterize the connectivity properties for each subregion. Finally, meta-based behavioral domain analysis was used to identify the main functions that each subregion participated in.The proposed parcellation scheme for BA46 with multimodal neuroimaging technique revealed that the rostral ventral BA46(BA46rv) mainly participated in interoception and somesthesis. The rostral dorsal BA46(BA46rd) primarily participated in visual and spatial perception, social cognition, and default model network. The posterior dorsal BA46(BA46pd) was mainly involved in execution control and attention; whereas the posterior ventral BA46(BA46pv) mainly participated in somesthesis perception and spatial working memory. These findings indicated that the parcellation results for BA46 were rational and reliable and demonstrated that BA46 is not a functional humongous area, but has different functional subregions.