Neural Data Sciences
The neural data sciences research realm involves using novel methods to model brain function and brain-system dysfunction; to interpret multimodal brain measurements; and to make health diagnoses from potentially large and multimodal datasets. Research activities in this category include imaging, conducting unit recordings, merging multi-modal data streams, and analyzing and linking models to data through assimilation, machine learning, or other methods. Researchers use computation, modeling, statistics, neuroimaging, electrophysiology, behavioral and disease modeling, machine learning, and artificial intelligence to accomplish their research goals.
Research examples:
- Signatures of arousal from functional MRI and electroencephalogram measurements (Xiao Liu)
- Novel machine learning approaches using Koopman operator methods to model and control big data (Bruce Gluckman)
- Cloud-based model predictions of concussion-related brain damage in sports from mouth-worn accelerometers (Reuban Kraft)
- Neuroimaging approaches to understanding differentiation of Parkinson’s disease clinical trajectories (Xuemei Huang)
- Analysis of the concurrent multimodal measures and the interactions between behavior, neural activity, and hemodynamics (Patrick Drew)
- Modeling the brain network organization and reorganization under different modulatory conditions (Nanyin Zhang)
- Statistical frameworks and machine learning algorithms for predicting differences in neurological disease phenotypes from multimodal data (Runze Li)
