CMStatistics 2016: Start Registration
View Submission - CMStatistics
B1234
Title: A spatio-temporal nonparametric Bayesian model of multi-subject fMRI data Authors:  Michele Guindani - University of California, Irvine (United States) [presenting]
Marina Vannucci - Rice University (United States)
Abstract: A unified, probabilistically coherent framework is proposed for the analysis of task-related brain activity in multi-subject fMRI experiments. This is distinct from two-stage group analysis approaches traditionally considered in the fMRI literature, which separate the inference on the individual fMRI time courses from the inference at the population level. In our modeling approach we consider a spatio-temporal linear regression model and specifically account for the between-subjects heterogeneity in neuronal activity via a spatially informed multi-subject nonparametric variable selection prior. For posterior inference, in addition to Markov chain Monte Carlo sampling algorithms, we develop suitable Variational Bayes algorithms. We show on simulated data that variational Bayes inference achieves satisfactory results at a much reduced computational costs than using MCMC, allowing scalability of our methods. In an application to data collected to assess brain responses to emotional stimuli our method correctly detects activation in visual areas when visual stimuli are presented.