Multivariate fusion of EEG and functional MRI data using ICA: Algorithm choice and performance analysis

Yuri Levin-Schwartz; Vince D. Calhoun; Tülay Adalı

doi:10.1007/978-3-319-22482-4_57

Multivariate fusion of EEG and functional MRI data using ICA: Algorithm choice and performance analysis

Yuri Levin-Schwartz, Vince D. Calhoun, Tülay Adalı

School of Medicine

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

Abstract

It has become common for neurological studies to gather data from multiple modalities, since the modalities examine complementary aspects of neural activity. Functional magnetic resonance imaging (fMRI) and electroencephalogram (EEG) data, in particular, enable the study of functional changes within the brain at different temporal and spatial scales; hence their fusion has received much attention. Joint independent component analysis (jICA) enables symmetric and fully multivariate fusion of these modalities and is thus one of the most widely used methods. In its application to jICA, Infomax has been the widely used, however the relative performance of Infomax is rarely shown on real neurological data, since the ground truth is not known. We propose the use of number of voxels in physically meaningful masks and statistical significance to assess algorithm performance of ICA for jICA on real data and show that entropy bound minimization (EBM) provides a more attractive solution for jICA of EEG and fMRI.

Original language	English (US)
Title of host publication	Latent Variable Analysis and Signal Separation - 12th International Conference, LVA/ICA 2015, Proceedings
Editors	Zbynĕk Koldovský, Emmanuel Vincent, Arie Yeredor, Petr Tichavský
Publisher	Springer Verlag
Pages	489-496
Number of pages	8
ISBN (Print)	9783319224817
DOIs	https://doi.org/10.1007/978-3-319-22482-4_57
State	Published - 2015
Event	12th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICA 2015 - Liberec, Czech Republic Duration: Aug 25 2015 → Aug 28 2015

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9237
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	12th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICA 2015
Country/Territory	Czech Republic
City	Liberec
Period	8/25/15 → 8/28/15

Keywords

Data fusion
EEG
FMRI
Independent component analysis
Medical imaging

ASJC Scopus subject areas

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-319-22482-4_57

Cite this

Levin-Schwartz, Y., Calhoun, V. D., & Adalı, T. (2015). Multivariate fusion of EEG and functional MRI data using ICA: Algorithm choice and performance analysis. In Z. Koldovský, E. Vincent, A. Yeredor, & P. Tichavský (Eds.), Latent Variable Analysis and Signal Separation - 12th International Conference, LVA/ICA 2015, Proceedings (pp. 489-496). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9237). Springer Verlag. https://doi.org/10.1007/978-3-319-22482-4_57

Multivariate fusion of EEG and functional MRI data using ICA: Algorithm choice and performance analysis. / Levin-Schwartz, Yuri; Calhoun, Vince D.; Adalı, Tülay.
Latent Variable Analysis and Signal Separation - 12th International Conference, LVA/ICA 2015, Proceedings. ed. / Zbynĕk Koldovský; Emmanuel Vincent; Arie Yeredor; Petr Tichavský. Springer Verlag, 2015. p. 489-496 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9237).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Levin-Schwartz, Y, Calhoun, VD & Adalı, T 2015, Multivariate fusion of EEG and functional MRI data using ICA: Algorithm choice and performance analysis. in Z Koldovský, E Vincent, A Yeredor & P Tichavský (eds), Latent Variable Analysis and Signal Separation - 12th International Conference, LVA/ICA 2015, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9237, Springer Verlag, pp. 489-496, 12th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICA 2015, Liberec, Czech Republic, 8/25/15. https://doi.org/10.1007/978-3-319-22482-4_57

Levin-Schwartz Y, Calhoun VD, Adalı T. Multivariate fusion of EEG and functional MRI data using ICA: Algorithm choice and performance analysis. In Koldovský Z, Vincent E, Yeredor A, Tichavský P, editors, Latent Variable Analysis and Signal Separation - 12th International Conference, LVA/ICA 2015, Proceedings. Springer Verlag. 2015. p. 489-496. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-22482-4_57

Levin-Schwartz, Yuri ; Calhoun, Vince D. ; Adalı, Tülay. / Multivariate fusion of EEG and functional MRI data using ICA : Algorithm choice and performance analysis. Latent Variable Analysis and Signal Separation - 12th International Conference, LVA/ICA 2015, Proceedings. editor / Zbynĕk Koldovský ; Emmanuel Vincent ; Arie Yeredor ; Petr Tichavský. Springer Verlag, 2015. pp. 489-496 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{ea18374b0adb4b398ea56178ea769201,

title = "Multivariate fusion of EEG and functional MRI data using ICA: Algorithm choice and performance analysis",

abstract = "It has become common for neurological studies to gather data from multiple modalities, since the modalities examine complementary aspects of neural activity. Functional magnetic resonance imaging (fMRI) and electroencephalogram (EEG) data, in particular, enable the study of functional changes within the brain at different temporal and spatial scales; hence their fusion has received much attention. Joint independent component analysis (jICA) enables symmetric and fully multivariate fusion of these modalities and is thus one of the most widely used methods. In its application to jICA, Infomax has been the widely used, however the relative performance of Infomax is rarely shown on real neurological data, since the ground truth is not known. We propose the use of number of voxels in physically meaningful masks and statistical significance to assess algorithm performance of ICA for jICA on real data and show that entropy bound minimization (EBM) provides a more attractive solution for jICA of EEG and fMRI.",

keywords = "Data fusion, EEG, FMRI, Independent component analysis, Medical imaging",

author = "Yuri Levin-Schwartz and Calhoun, {Vince D.} and T{\"u}lay Adalı",

note = "Funding Information: Y. Levin-Schwartz—This work was supported in part by NSF grant NSF-IIS 1017718 and NIH grant R01 EB 005846. Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2015.; 12th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICA 2015 ; Conference date: 25-08-2015 Through 28-08-2015",

year = "2015",

doi = "10.1007/978-3-319-22482-4_57",

language = "English (US)",

isbn = "9783319224817",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "489--496",

editor = "Zbynĕk Koldovsk{\'y} and Emmanuel Vincent and Arie Yeredor and Petr Tichavsk{\'y}",

booktitle = "Latent Variable Analysis and Signal Separation - 12th International Conference, LVA/ICA 2015, Proceedings",

}

TY - GEN

T1 - Multivariate fusion of EEG and functional MRI data using ICA

T2 - 12th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICA 2015

AU - Levin-Schwartz, Yuri

AU - Calhoun, Vince D.

AU - Adalı, Tülay

N1 - Funding Information: Y. Levin-Schwartz—This work was supported in part by NSF grant NSF-IIS 1017718 and NIH grant R01 EB 005846. Publisher Copyright: © Springer International Publishing Switzerland 2015.

PY - 2015

Y1 - 2015

N2 - It has become common for neurological studies to gather data from multiple modalities, since the modalities examine complementary aspects of neural activity. Functional magnetic resonance imaging (fMRI) and electroencephalogram (EEG) data, in particular, enable the study of functional changes within the brain at different temporal and spatial scales; hence their fusion has received much attention. Joint independent component analysis (jICA) enables symmetric and fully multivariate fusion of these modalities and is thus one of the most widely used methods. In its application to jICA, Infomax has been the widely used, however the relative performance of Infomax is rarely shown on real neurological data, since the ground truth is not known. We propose the use of number of voxels in physically meaningful masks and statistical significance to assess algorithm performance of ICA for jICA on real data and show that entropy bound minimization (EBM) provides a more attractive solution for jICA of EEG and fMRI.

AB - It has become common for neurological studies to gather data from multiple modalities, since the modalities examine complementary aspects of neural activity. Functional magnetic resonance imaging (fMRI) and electroencephalogram (EEG) data, in particular, enable the study of functional changes within the brain at different temporal and spatial scales; hence their fusion has received much attention. Joint independent component analysis (jICA) enables symmetric and fully multivariate fusion of these modalities and is thus one of the most widely used methods. In its application to jICA, Infomax has been the widely used, however the relative performance of Infomax is rarely shown on real neurological data, since the ground truth is not known. We propose the use of number of voxels in physically meaningful masks and statistical significance to assess algorithm performance of ICA for jICA on real data and show that entropy bound minimization (EBM) provides a more attractive solution for jICA of EEG and fMRI.

KW - Data fusion

KW - EEG

KW - FMRI

KW - Independent component analysis

KW - Medical imaging

UR - http://www.scopus.com/inward/record.url?scp=84944672872&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84944672872&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-22482-4_57

DO - 10.1007/978-3-319-22482-4_57

M3 - Conference contribution

AN - SCOPUS:84944672872

SN - 9783319224817

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 489

EP - 496

BT - Latent Variable Analysis and Signal Separation - 12th International Conference, LVA/ICA 2015, Proceedings

A2 - Koldovský, Zbynĕk

A2 - Vincent, Emmanuel

A2 - Yeredor, Arie

A2 - Tichavský, Petr

PB - Springer Verlag

Y2 - 25 August 2015 through 28 August 2015

ER -

Multivariate fusion of EEG and functional MRI data using ICA: Algorithm choice and performance analysis

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this