Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The Statistical Analysis of Functional MRI DataBayesian Methods in fMRI

The Statistical Analysis of Functional MRI Data: Bayesian Methods in fMRI In some ways the Bayesian framework is ideal for the analysis of functional MRI data. As we have seen in previous chapters, the data are often described in a hierarchical manner, with voxel-level models being embedded in subject- level models, which in turn may be nested in a group-level model. The hier- archical nature of the standard general linear model approach fits well into the Bayesian setting (Friston et al., 2002). Spatiotemporal models are an- other class that well describe functional neuroimaging data, and these too lend themselves quite naturally to a Bayesian analysis. Indeed, in Chapter 6 we saw several examples of Bayesian spatial or spatiotemporal analyses (for example, G¨ ossl et al. 2000; Hartvig and Jensen 2000; G¨ ossl et al. 2001; Smith et al. 2003). In addition, the basis function approaches that incor- porate anatomical information, as discussed in Section 8.2, have a distinctly Bayesian “flavor” even if they aren’t explicitly Bayes methods. The well-known criticisms of classical significance testing – the non- intuitive meaning of a p-value, the lack of symmetry between the null and alternative hypotheses (such that the null can never be accepted), increas- ing sensitivity with sample size so that a “statistically http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png

The Statistical Analysis of Functional MRI DataBayesian Methods in fMRI

Download PDF
13 pages

Loading next page...
 
/lp/springer-journals/the-statistical-analysis-of-functional-mri-data-bayesian-methods-in-9uGhGXO0kZ
Publisher
Springer New York
Copyright
© Springer-Verlag New York 2008
ISBN
978-0-387-78190-7
Pages
1 –13
DOI
10.1007/978-0-387-78191-4_9
Publisher site
See Chapter on Publisher Site

Abstract

In some ways the Bayesian framework is ideal for the analysis of functional MRI data. As we have seen in previous chapters, the data are often described in a hierarchical manner, with voxel-level models being embedded in subject- level models, which in turn may be nested in a group-level model. The hier- archical nature of the standard general linear model approach fits well into the Bayesian setting (Friston et al., 2002). Spatiotemporal models are an- other class that well describe functional neuroimaging data, and these too lend themselves quite naturally to a Bayesian analysis. Indeed, in Chapter 6 we saw several examples of Bayesian spatial or spatiotemporal analyses (for example, G¨ ossl et al. 2000; Hartvig and Jensen 2000; G¨ ossl et al. 2001; Smith et al. 2003). In addition, the basis function approaches that incor- porate anatomical information, as discussed in Section 8.2, have a distinctly Bayesian “flavor” even if they aren’t explicitly Bayes methods. The well-known criticisms of classical significance testing – the non- intuitive meaning of a p-value, the lack of symmetry between the null and alternative hypotheses (such that the null can never be accepted), increas- ing sensitivity with sample size so that a “statistically

Published: Jun 7, 2008

Keywords: Bayesian Method; fMRI Data; Reversible Jump; Marked Point Process; fMRI Data Analysis

There are no references for this article.