Dimension reduction for individual ICA to decompose FMRI during real-world experiences: Principal component analysis vs. canonical correlation analysis

Valeri Tsatsishvili; Fengyu Cong; Tuomas Puoliväli; Vinoo Alluri; Petri Toiviainen; Asoke K. Nandi; Elvira Brattico; Tapani Ristaniemi

Dimension reduction for individual ICA to decompose FMRI during real-world experiences: Principal component analysis vs. canonical correlation analysis

Valeri Tsatsishvili, Fengyu Cong, Tuomas Puoliväli, Vinoo Alluri, Petri Toiviainen, Asoke K. Nandi, Elvira Brattico, Tapani Ristaniemi

Institut for Klinisk Medicin - Center for Music in the Brain

Publikation: Bidrag til bog/antologi/rapport/proceeding › Konferencebidrag i proceedings › Forskning › peer review

2 Citationer (Scopus)

Abstract

Group independent component analysis (ICA) with special assumptions is often used for analyzing functional magnetic resonance imaging (fMRI) data. Before ICA, dimension reduction is applied to separate signal and noise subspaces. For analyzing noisy fMRI data of individual participants in free-listening to naturalistic and long music, we applied individual ICA and therefore avoided the assumptions of Group ICA. We also compared principal component analysis (PCA) and canonical correlation analysis (CCA) for dimension reduction of such fMRI data. We found interesting brain activity associated with music across majority of participants, and found that PCA and CCA were comparable for dimension reduction.

Originalsprog	Engelsk
Titel	ESANN 2013 proceedings, 21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Antal sider	6
Publikationsdato	2013
Sider	137-142
ISBN (Trykt)	9782874190810
Status	Udgivet - 2013
Begivenhed	21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2013 - Bruges, Belgien Varighed: 24 apr. 2013 → 26 apr. 2013

Konference

Konference	21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2013
Land/Område	Belgien
By	Bruges
Periode	24/04/2013 → 26/04/2013

Andre filer og links

http://www.scopus.com/inward/record.url?scp=84887066666&partnerID=8YFLogxK

Citationsformater

Tsatsishvili, V., Cong, F., Puoliväli, T., Alluri, V., Toiviainen, P., Nandi, A. K., Brattico, E., & Ristaniemi, T. (2013). Dimension reduction for individual ICA to decompose FMRI during real-world experiences: Principal component analysis vs. canonical correlation analysis. I ESANN 2013 proceedings, 21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (s. 137-142)

@inproceedings{0939d8e448a74f8b8b444451416b1c49,

title = "Dimension reduction for individual ICA to decompose FMRI during real-world experiences: Principal component analysis vs. canonical correlation analysis",

abstract = "Group independent component analysis (ICA) with special assumptions is often used for analyzing functional magnetic resonance imaging (fMRI) data. Before ICA, dimension reduction is applied to separate signal and noise subspaces. For analyzing noisy fMRI data of individual participants in free-listening to naturalistic and long music, we applied individual ICA and therefore avoided the assumptions of Group ICA. We also compared principal component analysis (PCA) and canonical correlation analysis (CCA) for dimension reduction of such fMRI data. We found interesting brain activity associated with music across majority of participants, and found that PCA and CCA were comparable for dimension reduction.",

author = "Valeri Tsatsishvili and Fengyu Cong and Tuomas Puoliv{\"a}li and Vinoo Alluri and Petri Toiviainen and Nandi, {Asoke K.} and Elvira Brattico and Tapani Ristaniemi",

year = "2013",

language = "English",

isbn = "9782874190810",

pages = "137--142",

booktitle = "ESANN 2013 proceedings, 21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning",

note = "21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2013 ; Conference date: 24-04-2013 Through 26-04-2013",

}

Tsatsishvili, V, Cong, F, Puoliväli, T, Alluri, V, Toiviainen, P, Nandi, AK, Brattico, E & Ristaniemi, T 2013, Dimension reduction for individual ICA to decompose FMRI during real-world experiences: Principal component analysis vs. canonical correlation analysis. i ESANN 2013 proceedings, 21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. s. 137-142, 21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2013, Bruges, Belgien, 24/04/2013.

Dimension reduction for individual ICA to decompose FMRI during real-world experiences: Principal component analysis vs. canonical correlation analysis. / Tsatsishvili, Valeri; Cong, Fengyu; Puoliväli, Tuomas et al.
ESANN 2013 proceedings, 21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. 2013. s. 137-142.

Publikation: Bidrag til bog/antologi/rapport/proceeding › Konferencebidrag i proceedings › Forskning › peer review

TY - GEN

T1 - Dimension reduction for individual ICA to decompose FMRI during real-world experiences

T2 - 21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2013

AU - Tsatsishvili, Valeri

AU - Cong, Fengyu

AU - Puoliväli, Tuomas

AU - Alluri, Vinoo

AU - Toiviainen, Petri

AU - Nandi, Asoke K.

AU - Brattico, Elvira

AU - Ristaniemi, Tapani

PY - 2013

Y1 - 2013

N2 - Group independent component analysis (ICA) with special assumptions is often used for analyzing functional magnetic resonance imaging (fMRI) data. Before ICA, dimension reduction is applied to separate signal and noise subspaces. For analyzing noisy fMRI data of individual participants in free-listening to naturalistic and long music, we applied individual ICA and therefore avoided the assumptions of Group ICA. We also compared principal component analysis (PCA) and canonical correlation analysis (CCA) for dimension reduction of such fMRI data. We found interesting brain activity associated with music across majority of participants, and found that PCA and CCA were comparable for dimension reduction.

AB - Group independent component analysis (ICA) with special assumptions is often used for analyzing functional magnetic resonance imaging (fMRI) data. Before ICA, dimension reduction is applied to separate signal and noise subspaces. For analyzing noisy fMRI data of individual participants in free-listening to naturalistic and long music, we applied individual ICA and therefore avoided the assumptions of Group ICA. We also compared principal component analysis (PCA) and canonical correlation analysis (CCA) for dimension reduction of such fMRI data. We found interesting brain activity associated with music across majority of participants, and found that PCA and CCA were comparable for dimension reduction.

UR - http://www.scopus.com/inward/record.url?scp=84887066666&partnerID=8YFLogxK

M3 - Article in proceedings

AN - SCOPUS:84887066666

SN - 9782874190810

SP - 137

EP - 142

BT - ESANN 2013 proceedings, 21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

Y2 - 24 April 2013 through 26 April 2013

ER -

Tsatsishvili V, Cong F, Puoliväli T, Alluri V, Toiviainen P, Nandi AK et al. Dimension reduction for individual ICA to decompose FMRI during real-world experiences: Principal component analysis vs. canonical correlation analysis. I ESANN 2013 proceedings, 21st European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. 2013. s. 137-142

Dimension reduction for individual ICA to decompose FMRI during real-world experiences: Principal component analysis vs. canonical correlation analysis

Abstract

Konference

Andre filer og links

Fingeraftryk

Citationsformater