Toward a standardized structural–functional group connectome in MNI space

• Published in: NeuroImage (Orlando, Fla.) Vol. 124; no. Pt A; pp. 310 - 322
• Main Authors: Horn, Andreas, Blankenburg, Felix
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.01.2016; Elsevier Limited
• Subjects: Adult; Algorithms; Brain; Brain - anatomy & histology; Connectome; Connectome - methods; Datasets; Diffusion Tensor Imaging - methods; dMRI; DTI; Experiments; Female; fMRI; Functional parcellation of the thalamus; Humans; Image Processing, Computer-Assisted; Magnetic Resonance Imaging - methods; Male; Middle Aged; MNI-space; Neural Pathways - anatomy & histology; Resting-state; Software; Structure–function agreement; Studies; Voxel-wise connectivity; White Matter - anatomy & histology; fMRI; Atlas; Functional parcellation of the thalamus; DTI; Resting-state; Structure–function agreement; Connectome; MNI-space; Voxel-wise connectivity; dMRI
• ISSN: 1053-8119; 1095-9572
• DOI: 10.1016/j.neuroimage.2015.08.048

The analysis of the structural architecture of the human brain in terms of connectivity between its subregions has provided profound insights into its underlying functional organization and has coined the concept of the “connectome”, a structural description of the elements forming the human brain and the connections among them. Here, as a proof of concept, we introduce a novel group connectome in standard space based on a large sample of 169 subjects from the Enhanced Nathan Kline Institute-Rockland Sample (eNKI-RS). Whole brain structural connectomes of each subject were estimated with a global tracking approach, and the resulting fiber tracts were warped into standard stereotactic (MNI) space using DARTEL. Employing this group connectome, the results of published tracking studies (i.e., the JHU white matter and Oxford thalamic connectivity atlas) could be largely reproduced directly within MNI space. In a second analysis, a study that examined structural connectivity between regions of a functional network, namely the default mode network, was reproduced. Voxel-wise structural centrality was then calculated and compared to others' findings. Furthermore, including additional resting-state fMRI data from the same subjects, structural and functional connectivity matrices between approximately forty thousand nodes of the brain were calculated. This was done to estimate structure–function agreement indices of voxel-wise whole brain connectivity. Taken together, the combination of a novel whole brain fiber tracking approach and an advanced normalization method led to a group connectome that allowed (at least heuristically) performing fiber tracking directly within MNI space. Such an approach may be used for various purposes like the analysis of structural connectivity and modeling experiments that aim at studying the structure–function relationship of the human connectome. Moreover, it may even represent a first step toward a standard DTI template of the human brain in stereotactic space. The standardized group connectome might thus be a promising new resource to better understand and further analyze the anatomical architecture of the human brain on a population level. •Structural group connectome in standard stereotactic (MNI) space•Open data release of voxel-wise structural and functional connectivity matrices•Large cohort of subjects with an age distribution ranging across the whole lifespan