Baba C. Vemuri, Ph.D., the Wilson and Marie Collins Professor in Engineering, and his team at the Department of Computer & Information Science & Engineering (CISE) recently received the best poster award at the biennial International Conference on Information Processing in Medical Imaging for research in processing advanced magnetic resonance imaging (MRI) data.
The research paper, titled “A Higher Order Manifold-valued Convolutional Neural Network with Applications in Diffusion MRI Processing,” was co-authored by Jose Bouza, a CISE master’s student; Chun-Hao Yang, a Ph.D. student in the UF Department of Statistics; David Vaillancourt, Ph.D., professor and chair from the UF Department of Applied Physiology & Kinesiology; and Dr. Vemuri. The award was received by Dr. Vemuri’s graduate student, Jose Bouza.
“This research will make advanced MRI techniques such as the dMRI clinically viable and hence benefit the treatment of Neurological Disorders such as Parkinson’s, Alzheimer’s and others,” Dr. Vemuri said.
“The research in this paper addresses the fundamental problem of reducing the data acquisition and reconstruction time for diffusion magnetic resonance images (dMRI),” Dr. Vemuri said. “dMRI is a non-invasive diagnostic imaging technique that allows one to probe the central nervous system in the human body and infer any irregularities in nerve fiber patterns. We present a novel geometric AI deep-learning technique that can successfully reduce the dMRI reconstruction time from hours to a few minutes.”
This research explores how to make dMRI a more effective diagnostic imaging tool. The team plans to implement their reconstruction algorithms on a clinical scanner and continue their research in the field of Geometric deep learning with applications to Neurology and Neuroscience.
Allison Logan
Marketing & Communications Specialist
Herbert Wertheim College of Engineering