This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 15
Type: Topic Contributed
Date/Time: Sunday, August 1, 2010 : 2:00 PM to 3:50 PM
Sponsor: Biometrics Section
Abstract - #307637
Title: A Probabilistic Group Independent Component Analysis Model and a Fast Approximate Estimation Approach
Author(s): Ying Guo*+
Companies: Emory University
Address: 1518 Clifton RD, Atlanta, GA, 30322,
Keywords: Group independent component analysis ; functional magnetic resonance imaging (fMRI) ; maximum likelihood method ; EM algorithm ; factorized variational approximation
Abstract:

Independent component analysis (ICA) is a powerful tool for analyzing data from functional magnetic resonance imaging (fMRI) studies. In this talk, we present a general probabilistic ICA model for multi-subject fMRI data that can accommodate various group structures of spatio-temporal processes. A maximum likelihood method is developed for the proposed model. We propose a modified EM algorithm to obtain the ML estimates. One issue with our EM algorithm is that its computation increases exponentially with the number of independent components. Consequently, the estimation process is time-consuming when a large number of components are extracted. For faster estimation, we propose a factorized variational approximation of the EM algorithm. We compare the performance of the exact estimation and the fast approximation method through simulation studies and an fMRI data example.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2010 program




2010 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.