A semiparametric PCA approach to fMRI data analysis

Functional Magnetic Resonance (fMRI) data is most often analyzed using linear regression type methods that consider each voxel separately or by using exploratory methods such as Principal Component Analysis (PCA) or Independent Component Analysis (ICA). In this paper we introduce a model, which we c...

Full description

Saved in:
Bibliographic Details
Main Authors: Ulfarsson, M O, Solo, V
Format: Conference Proceeding
Language:English
Subjects:
Bayesian methods
Covariance matrix
Data analysis
functionalMagnetic Resonance Imaging (fMRI)
Independent component analysis
Low-frequency noise
Magnetic analysis
Magnetic resonance
Maximum likelihood estimation
Principal component analysis
Principal Component Analysis (PCA)
Regression
Signal analysis
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Functional Magnetic Resonance (fMRI) data is most often analyzed using linear regression type methods that consider each voxel separately or by using exploratory methods such as Principal Component Analysis (PCA) or Independent Component Analysis (ICA). In this paper we introduce a model, which we call XnPCA, that combines regression with PCA. Unlike the linear regression methods XnPCA allows for non-stationary noise. Additionally, since XnPCA is based on the maximum likelihood framework the Bayesian information criterion (BIC) can be used for model selection and comparison. We compare XnPCA to a regression model commonly used in fMRI research using real data from a combined visual-motor experiment.
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2010.5495164