Meeting Banner
Abstract #3038

Unsupervised multi-tissue decomposition of single-shell diffusion-weighted imaging by generalization to multi-modal data

Daan Christiaens1,2, Frederik Maes1,2, Stefan Sunaert2,3, and Paul Suetens1,2,4

1ESAT/PSI, Department of Electrical Engineering, KU Leuven, Leuven, Belgium, 2Medical Imaging Research Center, UZ Leuven, Leuven, Belgium, 3Translational MRI, Department of Imaging & Pathology, KU Leuven, Leuven, Belgium, 4Medical IT Department, iMinds, Leuven, Belgium

Recent method developments have improved the reconstruction of fibre orientation distributions in white matter by incorporating partial voluming with isotropic grey matter and CSF. Yet, their use is limited to multi-shell data. Here, we present a generalization of convexity-constrained non-negative spherical factorization to multi-modal data, and illustrate its use for decomposing single-shell diffusion-weighted data and T1 anatomical data in three tissue components. Results show that we can effectively reconstruct fibre orientation distributions and separate isotropic volume fractions of grey matter and CSF in single-shell data, even at low b-values.

This abstract and the presentation materials are available to members only; a login is required.

Join Here