Correspondence between Structure and Function in the Human Brain at Rest
Judith Maxine Segall, Elena A Allen, Rex E Jung, Erik B Erhardt, Sunil Kumar Arja, Kent A Kiehl, Vince D Calhoun
Frontiers in Neuroinformatics
Accepted 12 Mar 2012, 6:10
To further the understanding of basic and complex cognitive functions of the human brain, multidisciplinary neuroimaging research has explored both functional and structural connectivity. For structural connectivity, the most prevalent method has been diffusion weighted imaging, which measures the connections of large white matter bundles. Recently, functional connectivity has been measured using resting-state fMRI (rs-fMRI). Surprisingly, few studies have examined structural gray matter, which supports the BOLD response. The overall aim of this study is to explore how gray matter (GM) structure corresponds to function. A cohort of 603 healthy participants was scanned on the same 3T scanner at the Mind Research Network to investigate the spatial correlations between structure and function. This was done by applying spatial independent component analysis (ICA) to GMD maps, to delineate structural components based on the covariation of GMD between regions, and to rs-fMRIdata, to discover spatial patterns with common temporal features. Decomposed structural and functional components were then compared by spatial correlation. The basal ganglia network showed the highest structural to rs-functional component correlation (r=0.59). Our remaining results generally show correspondence between one structural network and several functional networks. We also studied relationships between the weights of different structural components and found networks in frontal and parietal regions showing covariation across subjects. We also identified the precuneus as a hub for in structural network correlations. In addition, we analyzed relationships between component weights and age, concluding that age has an effect on structural components.