|
Activity Number:
|
132
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 3, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #305057 |
|
Title:
|
An Application of the Spatial Bayesian Variable Selection to fMRI Time-Series Data
|
|
Author(s):
|
Kuo-Jung Lee*+
|
|
Companies:
|
The University of Minnesota-Twin Cities
|
|
Address:
|
1041 29TH Avenue SE Apt. #C, Minneapolis, MN, 55414,
|
|
Keywords:
|
fMRI ; MCMC ; Bayesian
|
|
Abstract:
|
Frequentist methods assessing human brain activity using data from fMRI (functional magnetic resonance imaging) experiments rely on results from the theory of Gaussian random fields. Such methods have several limitations. Bayesian methods provide an alternative that bypasses the problem of multiplicity. We use a spatial Bayesian variable selection method to allow automatic detection of the activation of a voxel corresponding to a given stimulus by introducing indicator variables with an Ising prior. We propose a more efficient way to evaluate the posterior quantities using Markov chain Monte Carlo (MCMC).
|
