Individual variability in the anatomical distribution of nodes participating in rich club structural networks.

With recent advances in computational analyses of structural neuroimaging, it is possible to comprehensively map neural connectivity, i.e., the brain connectome. The architectural organization of the connectome is believed to play an important role in several biological processes. Central to the conformation of the connectome are connectivity hubs, which are likely to be organized in accordance with the rich club phenomenon, as evidenced by graph theory analyses of neural architecture. It is yet unclear whether rich club connectivity hubs are consistently organized in the same anatomical framework across healthy adults. We constructed the brain connectome from 43 healthy adults, based on T1-weighted and diffusion tensor MRI data. Probabilistic fiber tractography was used to evaluate connectivity between each possible pair of cortical anatomical regions of interest. Connectivity hubs were identified in accordance with the rich club phenomenon applied to binarized matrices, and the variability in frequency of hub participation was assessed node-wise across all subjects. The anatomical location of nodes participating in rich club networks was fairly consistent across subjects. The most common locations for rich club nodes were identified in integrative areas, such as the cingulate and pericingulate regions, medial aspect of the occipital areas and precuneus; or else, they were found in important and specialized brain regions (such as the oribitofrontal cortex, caudate, fusiform gyrus, and hippocampus). Marked anatomical consistency exists across healthy brains in terms of nodal participation and location of rich club networks. The consistency of connections between integrative areas and specialized brain regions highlights a fundamental connectivity pattern shared among healthy brains. We propose that approaching brain connectivity with this framework of anatomical consistencies may have clinical implications for early detection of individual variability.