Proposal Description

Modern technology produces data with complex structure (e.g., high-dimensional, non-Euclidean)
and statistical methods need to be developed to extract useful information from these data modalities in order to move the field of precision medicine forward. These new methods require developing innovative modeling strategies and numerical methods to apply these models to data generated from health research. This session will describe recent advances in developing such methods to advance medical research, particularly in mental health research. Dr. Garcia will speak on "Using EEG as a tool to personalized medicine in depression treatments." Dr. Park will speak on "Bayesian Estimation of Covariate Assisted Principal Regression for brain functional connectivity" and Dr. Ju will speak on "Dimension reduction in regression with functional connectivity predictors" This session is relevant because most available methods for precision medicine are using suboptimal statistical tools that were developed for data with simpler structures. Using advanced methods will allow a fuller extraction of the available information in complex data modalities and will accelerate progress in precision medicine.