Abstract #1680

Using GPUs to accelerate computational diffusion MRI: From microstructure estimation to tractography and connectomes

Moises Hernandez-Fernandez^1,2, Istvan Reguly^3,4, Saad Jbabdi¹, Mike Giles³, Stephen Smith¹, and Stamatios N. Sotiropoulos^1,5

¹Oxford Centre for Functional MRI of the Brain (FMRIB), University of Oxford, Oxford, United Kingdom, ²Section for Biomedical Image Analysis (SBIA), University of Pennsylvania, Philadelphia, PA, United States, ³Oxford e-Research Centre, University of Oxford, Oxford, United Kingdom, ⁴Faculty of Information Technology and Bionics, Pazmany Peter Catholic University, Budapest, Hungary, ⁵Sir Peter Mansfield Imaging Centre, School of Medicine, University of Nottingham, Nottingham, United Kingdom

The great potential of computational diffusion MRI (dMRI) relies on indirect inference of tissue microstructure and brain connections, as modelling and tractography frameworks map diffusion measurements to neuroanatomical features. This mapping however can be computationally expensive, particularly given the trend of increasing dataset sizes and/or the increased complexity in biophysical modelling. We present here a number of frameworks for accelerating dMRI computations using Graphics Processing Units (GPUs), for both microstructure estimation and tractography/connectome generation. We show that despite differences in challenges for parallelising these problems, GPU-based designs can offer accelerations of more than two orders of magnitude.

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

Join Here