Diffusion MRI (dMRI) data obtained from a 7T scanner has novel and improved microstructural tissue information missing from data acquired on 3T scanners. In this work, we propose to use deep Convolutional Neural Networks (CNN) that use rotation invariant spherical harmonic (RISH) features to map the dMRI data (the raw signal) between scanners without changing the fiber orientation. We validate our algorithm on 40 Human Connectome Project (HCP) subjects with scans on both 3T and 7T (10 training + 30 test). Our preliminary results on 30 test subjects shows that CNN can indeed reliably obtain 7T dMRI data quality from 3T scans.
This abstract and the presentation materials are available to members only; a login is required.