Independent component analysis for brain fMRI does not select for independence

I. Daubechies; E. Roussos; S. Takerkart; M. Benharrosh; C. Golden; K. D'Ardenne; W. Richter; J. D. Cohen; J. Haxby

doi:10.1073/pnas.0903525106

Independent component analysis for brain fMRI does not select for independence

I. Daubechies, E. Roussos, S. Takerkart, M. Benharrosh, C. Golden, K. D'Ardenne, W. Richter, J. D. Cohen, J. Haxby

Research output: Contribution to journal › Article › peer-review

199 Scopus citations

Abstract

InfoMax and FastICA are the independent component analysis algorithms most used and apparently most effective for brain fMRI. We show that this is linked to their ability to handle effectively sparse components rather than independent components as such. The mathematical design of better analysis tools for brain fMRI should thus emphasize other mathematical characteristics than independence.

Original language	English (US)
Pages (from-to)	10415-10422
Number of pages	8
Journal	Proceedings of the National Academy of Sciences of the United States of America
Volume	106
Issue number	26
DOIs	https://doi.org/10.1073/pnas.0903525106
State	Published - Jun 30 2009
Externally published	Yes

ASJC Scopus subject areas

General

Access to Document

10.1073/pnas.0903525106

Cite this

@article{9e0c1c882d004faf9b6a6c54a24f134a,

title = "Independent component analysis for brain fMRI does not select for independence",

abstract = "InfoMax and FastICA are the independent component analysis algorithms most used and apparently most effective for brain fMRI. We show that this is linked to their ability to handle effectively sparse components rather than independent components as such. The mathematical design of better analysis tools for brain fMRI should thus emphasize other mathematical characteristics than independence.",

author = "I. Daubechies and E. Roussos and S. Takerkart and M. Benharrosh and C. Golden and K. D'Ardenne and W. Richter and Cohen, {J. D.} and J. Haxby",

year = "2009",

month = jun,

day = "30",

doi = "10.1073/pnas.0903525106",

language = "English (US)",

volume = "106",

pages = "10415--10422",

journal = "Proceedings of the National Academy of Sciences of the United States of America",

issn = "0027-8424",

publisher = "National Academy of Sciences",

number = "26",

}

TY - JOUR

T1 - Independent component analysis for brain fMRI does not select for independence

AU - Daubechies, I.

AU - Roussos, E.

AU - Takerkart, S.

AU - Benharrosh, M.

AU - Golden, C.

AU - D'Ardenne, K.

AU - Richter, W.

AU - Cohen, J. D.

AU - Haxby, J.

PY - 2009/6/30

Y1 - 2009/6/30

N2 - InfoMax and FastICA are the independent component analysis algorithms most used and apparently most effective for brain fMRI. We show that this is linked to their ability to handle effectively sparse components rather than independent components as such. The mathematical design of better analysis tools for brain fMRI should thus emphasize other mathematical characteristics than independence.

AB - InfoMax and FastICA are the independent component analysis algorithms most used and apparently most effective for brain fMRI. We show that this is linked to their ability to handle effectively sparse components rather than independent components as such. The mathematical design of better analysis tools for brain fMRI should thus emphasize other mathematical characteristics than independence.

UR - http://www.scopus.com/inward/record.url?scp=67649811101&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67649811101&partnerID=8YFLogxK

U2 - 10.1073/pnas.0903525106

DO - 10.1073/pnas.0903525106

M3 - Article

C2 - 19556548

AN - SCOPUS:67649811101

SN - 0027-8424

VL - 106

SP - 10415

EP - 10422

JO - Proceedings of the National Academy of Sciences of the United States of America

JF - Proceedings of the National Academy of Sciences of the United States of America

IS - 26

ER -

Independent component analysis for brain fMRI does not select for independence

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this