The combination of univariate and multivariate method for fMRI data analysis

Author

Xia, Weiwei ; Yan, Lirong ; Zhou, Zongtan ; Liu, Yadong ; Hu, Dewen

Author_Institution

Dept. of Autom. Control, Nat. Univ. of Defense Technol., Changsa

Volume

3

fYear

2005

Firstpage

1568

Lastpage

1573

Abstract

A combined method of univariate and multivariate analysis is presented in this paper to give a new way for fMRI data analysis. The univariate single-frame approach, which detects activations evoked by a specific task and describes the temporal characteristics of activations without prior assumptions of hemodynamic response function (HRF), can be applied as the first processing step. While the multivariate methods, i.e., spatial and temporal independent component analyses (sICA and tICA), are then brought in to analyze the derived spatiotemporal activations in the regions of interest (ROIs). The ICAs can, in the combined approach, reveal the subtle spatial patterns of the regional activation areas

Keywords

biomedical MRI; brain; independent component analysis; spatiotemporal phenomena; activation temporal characteristics; evoked activation detection; fMRI data analysis; multivariate method; spatial independent component analysis; spatiotemporal activations; temporal independent component analysis; univariate single-frame approach; Analysis of variance; Data analysis; Image analysis; Independent component analysis; Magnetic analysis; Magnetic resonance imaging; Positron emission tomography; Principal component analysis; Spatiotemporal phenomena; Statistical analysis; Single-frame; Z-score; independent component analysis(ICA); region of interest(ROI); sICA; tICA;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks and Brain, 2005. ICNN&B '05. International Conference on

Conference_Location

Beijing

Print_ISBN

0-7803-9422-4

Type

conf

DOI

10.1109/ICNNB.2005.1614931

Filename

1614931