Presentation 2002/5/10
Model-Driven fMRI Data Analysis by Splitting Temporal Space to Subspaces
Shinobu MIZUTA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In order to find activated area in brain related to introduced models, a model-driven method for fMRI is proposed. In this method, temporal space which describes the measured time-course sequences is decomposed into model subspace and noise subspace, and the noise subspace is also decomposed into confound subspace, which means source signals of no-interest, and random noise subspace. The former decomposition is directly carried out using introduced models, and the latter is carried out by eigenvalue decomposition. Activated area is extracted by comparing the model subspace and the random noise subspace. From experiments for simulated data, usefulness of the proposed method is shown.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) fMRI / Subspace / General linear model / Principal component analysis / Eigenvalue decomposition
Paper # PRMU2002-10
Date of Issue

Conference Information
Committee PRMU
Conference Date 2002/5/10(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Model-Driven fMRI Data Analysis by Splitting Temporal Space to Subspaces
Sub Title (in English)
Keyword(1) fMRI
Keyword(2) Subspace
Keyword(3) General linear model
Keyword(4) Principal component analysis
Keyword(5) Eigenvalue decomposition
1st Author's Name Shinobu MIZUTA
1st Author's Affiliation Department of Systems Science, Graduate School of Informatics, Kyoto University:Brain Science Institute, RIKEN()
Date 2002/5/10
Paper # PRMU2002-10
Volume (vol) vol.102
Number (no) 55
Page pp.pp.-
#Pages 8
Date of Issue