An algorithm to estimate anatomical connectivity between brain regions using diffusion MRI

The study of anatomical connectivity is essential for interpreting functional MRI data, and establishing how brain areas are linked together into networks to support higher order functions. Diffusion weighted MR images (DWI) and tractography provide a unique noninvasive tool to explore the connectional architecture of the brain. The identification of anatomical circuits associated with a specific function can be better accomplished by the joint application of diffusion and functional MRI. In this paper, we propose a simple algorithm to identify the set of pathways between two regions of interest. The method is based upon running deterministic tractography from all possible starting positions in the brain, and selecting trajectories that intersect both regions. We compare results from single fibre tractography using diffusion tensor imaging, and multifibre tractography using reduced encoding Persistent Angular Structure (PAS) MRI, on standard DWI datasets from healthy human volunteers. Our results show that, in comparison with single fibre tractography, the multifibre technique reveal additional putative routes of connection. We demonstrate highly consistent results of the proposed technique over a cohort of 16 healthy subjects.