Presentation 2003/3/10
MEG current source estimation and Bayes prior
Masa-aki SATO, Taku YOSHIOKA, Shigeki KAJIWARA, Keisuke TOYAMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose a variational Bayes method for estimating current source distribution from observed MEG data. In the method, we introduce a hierarchical prior which assumes that the current sources are localized in the several brain area and the current distribution is continuous. The fMRI data can be also incorporated into the hierarchical prior, so that the reliability of the estimation is increased.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) MEG / Current source estimation / Variational Bayes method / Hierarchical prior / Model posterior probability
Paper # NC2002-148
Date of Issue

Conference Information
Committee NC
Conference Date 2003/3/10(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 Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) MEG current source estimation and Bayes prior
Sub Title (in English)
Keyword(1) MEG
Keyword(2) Current source estimation
Keyword(3) Variational Bayes method
Keyword(4) Hierarchical prior
Keyword(5) Model posterior probability
1st Author's Name Masa-aki SATO
1st Author's Affiliation Human Information Science Labs, ATR International:CREST, Japan Science and Technology Corporation()
2nd Author's Name Taku YOSHIOKA
2nd Author's Affiliation NAIST:Human Information Science Labs.
3rd Author's Name Shigeki KAJIWARA
3rd Author's Affiliation Technology Research Lab., Shimadzu C0.
4th Author's Name Keisuke TOYAMA
4th Author's Affiliation Technology Research Lab., Shimadzu C0.
Date 2003/3/10
Paper # NC2002-148
Volume (vol) vol.102
Number (no) 729
Page pp.pp.-
#Pages 6
Date of Issue