Confronting the Forward and Inverse Problems in Functional Neuro-imaging: Enhancing the Connectivity Analysis of Brain Regions and the Prediction of Brain States

The goal of this research is to develop innovative methods to solve the forward and inverse problems in functional neuro-imaging:The forward problem: when a person performs a task or receives some stimuli, how do the activated regions of the brain interact with one another (connectivity analysis)?The inverse problem: given the neuro-imaging recordings of a person, how can his or her cognitive states be inferred from these data (brain state prediction)?The researchers strive to advance the state of the art in tackling these problems by:applying computationally intensive all-voxel-versus-all- voxel causality analysis;developing dimensionality reduction and data visualization techniques based on a new metric; andapplying these data reduction techniques to improve connectivity analysis and brain state prediction.