Researchers from the University of Oxford won the fifth annual Achievement Award for NVIDIA GPU Centers of Excellence in recognition of their work using GPUs to speed the analysis of the human brain’s underlying anatomical and structural organization.
The team, led by Mike Giles, received the top award, along with a $15,000 cash prize, for their work applying GPU acceleration to tractography, a 3D modeling technique for investigating human brain connections.
With GPUs handling the immense computing load, the team increased the speed of analysis of diffusion MRI (dMRI), a widely used method to reveal the wiring diagram of the brain and estimate tissue microstructures, by up to 150x (1 GPU v. 1 single-core CPU).
The Oxford team’s work is aimed at improving the investigation of neurological and psychiatric diseases — such as multiple sclerosis, schizophrenia and Alzheimer’s disease — and has the potential to improve the clinical applicability of dMRI.
A panel of experts selected three other teams from top universities as finalists from among our 23 GPU Centers of Excellence.
The finalists include:
University of Illinois, Urbana-Champaign, team led by Wen-mei Hwu, for their research on interactive molecular ray-tracing using GPUs, remote clouds and supercomputers, and virtual reality head-mounted displays.