TY - JOUR
T1 - aNy-way independent component analysis
AU - Duan, Kuaikuai
AU - Silva, Rogers F.
AU - Calhoun, Vince D.
AU - Liu, Jingyu
N1 - Publisher Copyright:
The copyright holder for this preprint is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission.
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2020/2/3
Y1 - 2020/2/3
N2 - Multimodal data fusion is a topic of great interest. Several fusion methods have been proposed to investigate coherent patterns and corresponding linkages across modalities, such as joint independent component analysis (jICA), multiset canonical correlation analysis (mCCA), mCCA+jICA, and parallel ICA. JICA exploits source independence but assumes shared loading parameters. MCCA maximizes correlation linkage across modalities directly but is limited to orthogonal features. While there is no theoretical limit to the number of modalities analyzed together by jICA, mCCA, or the two-step approach mCCA+jICA, these approaches require the same number of sources/components for all modalities. Parallel ICA, on the other hand, simultaneously maximizes correlation between modalities and independence of sources, while allowing different number of sources for each modality. However, only a very limited number of modalities and linkage pairs can be optimized. To overcome these limitations, we propose aNy-way ICA, a new model to simultaneously maximize the independence of sources and correlations across modalities. ANy-way ICA combines infomax ICA and Gaussian independent vector analysis (IVA-G) via a shared weight matrix model without orthogonality constraints. Simulation results demonstrate that aNy-way ICA can accurately recover sources and loadings, as well as the true covariance patterns, whether different modalities have the same or different number of sources. Moreover, aNy-way ICA outperforms mCCA and mCCA+jICA in terms of source and loading recovery accuracy, especially under noisy conditions.
AB - Multimodal data fusion is a topic of great interest. Several fusion methods have been proposed to investigate coherent patterns and corresponding linkages across modalities, such as joint independent component analysis (jICA), multiset canonical correlation analysis (mCCA), mCCA+jICA, and parallel ICA. JICA exploits source independence but assumes shared loading parameters. MCCA maximizes correlation linkage across modalities directly but is limited to orthogonal features. While there is no theoretical limit to the number of modalities analyzed together by jICA, mCCA, or the two-step approach mCCA+jICA, these approaches require the same number of sources/components for all modalities. Parallel ICA, on the other hand, simultaneously maximizes correlation between modalities and independence of sources, while allowing different number of sources for each modality. However, only a very limited number of modalities and linkage pairs can be optimized. To overcome these limitations, we propose aNy-way ICA, a new model to simultaneously maximize the independence of sources and correlations across modalities. ANy-way ICA combines infomax ICA and Gaussian independent vector analysis (IVA-G) via a shared weight matrix model without orthogonality constraints. Simulation results demonstrate that aNy-way ICA can accurately recover sources and loadings, as well as the true covariance patterns, whether different modalities have the same or different number of sources. Moreover, aNy-way ICA outperforms mCCA and mCCA+jICA in terms of source and loading recovery accuracy, especially under noisy conditions.
KW - ANy-way ICA
KW - Correlation
KW - Independence
KW - Multimodal data fusion
UR - http://www.scopus.com/inward/record.url?scp=85095661484&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85095661484&partnerID=8YFLogxK
U2 - 10.1101/2020.02.03.930156
DO - 10.1101/2020.02.03.930156
M3 - Article
AN - SCOPUS:85095661484
SN - 0309-1708
JO - Unknown Journal
JF - Unknown Journal
ER -