CMIC Seminar: Universal power-law scaling of water diffusion in human brain uncovers microstructural properties of the human white matter

Speaker: Jelle Veraart, iMinds Vision Lab, University of Antwerp, Center for Biomedical Imaging, New York University School of Medicine
UCL Contact: Dominique Drai (Visitors from outside UCL please email in advance).
Date/Time: 24 Aug 16, 13:00 - 14:00
Venue: Roberts 106

Abstract

Development of successful therapies for neurological disorders depends on our ability to diagnose and monitor the progression of brain pathologies at the cellular level. Physics and physiology limit the resolution of human MRI to millimeters, three orders of magnitude coarser than the cell dimensions of microns. Bridging the resolution gap can be achieved only indirectly, by virtue of biophysical modeling, which is meant to identify relevant microstructural parameters from the acquired MRI signal in every imaging voxel. The Achilles heel of model-based approaches is lack of direct validation of model assumptions — even histology cannot determine markers of cell integrity, such as diffusion coefficients of intra and extra-cellular water, and membrane permeability. Here, we identify a universal power-law scaling of the diffusion-weighted signal, up to b=10000 s/mm2, on a clinical scanner, a key to validate fundamental model assumptions for water diffusion in the brain. The specific value of the exponent α =1/2 is a signature of water confined within narrow “channels” (axons in the white matter), establishing that exchange between intra- and extra-axonal water is not relevant, and the fraction of fully restricted water is negligible. The observed scaling for the first time validates the overarching picture for microstructural modeling in the brain conjectured over the past decade, determines the information content of the unexpectedly large signal at high diffusion weightings, and bridges modern diffusion models with clinical MRI applications.