A Bayesian Model of Smoothing and Connectivity for Event-Related fMRI Time Series

• Activity Number: 31
• Type: Contributed
• Date/Time: Sunday, August 2, 2009 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistical Learning and Data Mining
• Author(s): Wesley K. Thompson and Dongli Zhou*+
• Companies: University of California, San Diego and University of Pittsburgh
• Keywords: Smoothing ; MCMC ; Bayesian ; Functional Connectivity ; fMRI ; time series

Abstract

Neuroscientists have become increasingly interested in exploring dynamic relationships among brain regions. The presence of such a relationship is denoted by the term "functional connectivity." We propose a methodology for exploring functional connectivity in event-related designs, where stimuli are presented at a sufficient separation to examine dynamic responses in multiple brain regions. Our methodology simultaneously determines the level of smoothing to obtain the underlying noise-free BOLD response and functional connectivity among several regions. Smoothing is accomplished through an empirical basis via functional principal components analysis. The coefficients of the basis are assumed to be correlated across regions, and the nature and strength of functional connectivity is derived from this correlation matrix. The method is implemented via Bayesian MCMC on an fMRI data set.