Learning from imaging data to model brain activity
Published in Brain Activity, Brain/Neurology, Scanners and Imaging.
In this article, we introduce VanDEEPol, a hybrid AI/mechanistic model to predict brain activity and structure from imaging data. The model significantly boosts predictive accuracy compared to previous methods. By predicting brain activity from relatively sparse imaging data, VanDEEPol may eventually help to detect medical disorders or design brain-computer interfaces.
Intricate interactions among billions of neurons occur constantly in our brains, underlying our thoughts, functions, and behaviors. With sophisticated techniques such as functional magnetic resonance imaging (fMRI) and calcium imaging (CaI), we can map these elaborate connections with unprecedented detail.
Algorithmic models that describe whole-brain activity based on these maps could yield clues into how our brains work, both normally and in disease states.