Georgia Tech Neuro Seminar Series
“Machine Learning Approaches to Characterizing and Interpreting Brain-wide Dynamics”
Shella Keilholz, Ph.D.
Wallace H. Coulter Department of Biomedical Engineering
Georgia Tech / Emory University
The intrinsic activity of the brain is organized into networks and motifs that vary over time. To understand how coordinated macroscale patterns of intrinsic activity flow across the brain’s structural network, new analysis approaches that can extract interpretable spatial and temporal features are required. We demonstrate the use of machine learning to extract prototypical spatiotemporal trajectories of brain activity that provide a sparse and elegant basis of representation. To better interpret the spatiotemporal patterns of brain activity, we incorporate brain network models as constraints, forcing the algorithm to learn parameters that can be measured empirically. This interpretable machine learning-based approach paves the way for multimodal experiments that can validate or disprove the accuracy of different brain network models.
About the Speaker
Shella Keilholz obtained her B.S. in Physics from the University of Missouri—Rolla (now Missouri University of Science and Technology). Her Ph.D. in Engineering Physics at the University of Virginia focused on quantitative measurements of perfusion with arterial spin labeling MRI. After graduation, she went to the NIH as a postodoctoral researcher in Dr. Alan Koretsky’s lab to learn functional neuroimaging. She is currently a professor in the joint Emory University/Georgia Tech Biomedical Engineering department. Her research seeks to elucidate the neurophysiological processes that underlie the BOLD signal and develop analytical techniques that leverage spatial and temporal information to separate contributions from different sources.
- Workflow Status: Published
- Created By: Colly Mitchell
- Created: 08/26/2021
- Modified By: Colly Mitchell
- Modified: 08/26/2021