Sparse infomax based on hoyer projection and its application to simulated structural MRI and SNP data

Kuaikuai Duan, Rogers F. Silva, Jiayu Chen, Dongdong Lin, Vince D. Calhoun, Jingyu Liu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Independent component analysis has been widely applied to brain imaging and genetic data analyses for its ability to identify interpretable latent sources. Nevertheless, leveraging source sparsity in a more granular way may further improve its ability to optimize the solution for certain data types. For this purpose, we propose a sparse infomax algorithm based on nonlinear Hoyer projection, leveraging both sparsity and statistical independence of latent sources. The proposed algorithm iteratively updates the unmixing matrix by infomax (for independence) and the sources by Hoyer projection (for sparsity), feeding the sparse sources back as input data for the next iteration. Consequently, sparseness propagates effectively through infomax iterations, producing sources with more desirable properties. Simulation results on both brain imaging and genetic data demonstrate that the proposed algorithm yields improved pattern recovery, particularly under low signal-to-noise ratio conditions, as well as improved sparseness compared to traditional infomax.

Original languageEnglish (US)
Title of host publicationISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging
PublisherIEEE Computer Society
Pages418-421
Number of pages4
ISBN (Electronic)9781538636411
DOIs
StatePublished - Apr 2019
Externally publishedYes
Event16th IEEE International Symposium on Biomedical Imaging, ISBI 2019 - Venice, Italy
Duration: Apr 8 2019Apr 11 2019

Publication series

NameProceedings - International Symposium on Biomedical Imaging
Volume2019-April
ISSN (Print)1945-7928
ISSN (Electronic)1945-8452

Conference

Conference16th IEEE International Symposium on Biomedical Imaging, ISBI 2019
CountryItaly
CityVenice
Period4/8/194/11/19

Fingerprint

Magnetic resonance imaging
Single Nucleotide Polymorphism
Neuroimaging
Brain
Imaging techniques
Independent component analysis
Signal-To-Noise Ratio
Signal to noise ratio
Recovery

Keywords

  • Hoyer projection
  • Imaging data
  • Pattern recovery
  • Snp data
  • Sparse infomax

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Duan, K., Silva, R. F., Chen, J., Lin, D., Calhoun, V. D., & Liu, J. (2019). Sparse infomax based on hoyer projection and its application to simulated structural MRI and SNP data. In ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging (pp. 418-421). [8759599] (Proceedings - International Symposium on Biomedical Imaging; Vol. 2019-April). IEEE Computer Society. https://doi.org/10.1109/ISBI.2019.8759599

Sparse infomax based on hoyer projection and its application to simulated structural MRI and SNP data. / Duan, Kuaikuai; Silva, Rogers F.; Chen, Jiayu; Lin, Dongdong; Calhoun, Vince D.; Liu, Jingyu.

ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society, 2019. p. 418-421 8759599 (Proceedings - International Symposium on Biomedical Imaging; Vol. 2019-April).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Duan, K, Silva, RF, Chen, J, Lin, D, Calhoun, VD & Liu, J 2019, Sparse infomax based on hoyer projection and its application to simulated structural MRI and SNP data. in ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging., 8759599, Proceedings - International Symposium on Biomedical Imaging, vol. 2019-April, IEEE Computer Society, pp. 418-421, 16th IEEE International Symposium on Biomedical Imaging, ISBI 2019, Venice, Italy, 4/8/19. https://doi.org/10.1109/ISBI.2019.8759599
Duan K, Silva RF, Chen J, Lin D, Calhoun VD, Liu J. Sparse infomax based on hoyer projection and its application to simulated structural MRI and SNP data. In ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society. 2019. p. 418-421. 8759599. (Proceedings - International Symposium on Biomedical Imaging). https://doi.org/10.1109/ISBI.2019.8759599
Duan, Kuaikuai ; Silva, Rogers F. ; Chen, Jiayu ; Lin, Dongdong ; Calhoun, Vince D. ; Liu, Jingyu. / Sparse infomax based on hoyer projection and its application to simulated structural MRI and SNP data. ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging. IEEE Computer Society, 2019. pp. 418-421 (Proceedings - International Symposium on Biomedical Imaging).
@inproceedings{ed4c94b4b6884300ab571afc728f0243,
title = "Sparse infomax based on hoyer projection and its application to simulated structural MRI and SNP data",
abstract = "Independent component analysis has been widely applied to brain imaging and genetic data analyses for its ability to identify interpretable latent sources. Nevertheless, leveraging source sparsity in a more granular way may further improve its ability to optimize the solution for certain data types. For this purpose, we propose a sparse infomax algorithm based on nonlinear Hoyer projection, leveraging both sparsity and statistical independence of latent sources. The proposed algorithm iteratively updates the unmixing matrix by infomax (for independence) and the sources by Hoyer projection (for sparsity), feeding the sparse sources back as input data for the next iteration. Consequently, sparseness propagates effectively through infomax iterations, producing sources with more desirable properties. Simulation results on both brain imaging and genetic data demonstrate that the proposed algorithm yields improved pattern recovery, particularly under low signal-to-noise ratio conditions, as well as improved sparseness compared to traditional infomax.",
keywords = "Hoyer projection, Imaging data, Pattern recovery, Snp data, Sparse infomax",
author = "Kuaikuai Duan and Silva, {Rogers F.} and Jiayu Chen and Dongdong Lin and Calhoun, {Vince D.} and Jingyu Liu",
year = "2019",
month = "4",
doi = "10.1109/ISBI.2019.8759599",
language = "English (US)",
series = "Proceedings - International Symposium on Biomedical Imaging",
publisher = "IEEE Computer Society",
pages = "418--421",
booktitle = "ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging",

}

TY - GEN

T1 - Sparse infomax based on hoyer projection and its application to simulated structural MRI and SNP data

AU - Duan, Kuaikuai

AU - Silva, Rogers F.

AU - Chen, Jiayu

AU - Lin, Dongdong

AU - Calhoun, Vince D.

AU - Liu, Jingyu

PY - 2019/4

Y1 - 2019/4

N2 - Independent component analysis has been widely applied to brain imaging and genetic data analyses for its ability to identify interpretable latent sources. Nevertheless, leveraging source sparsity in a more granular way may further improve its ability to optimize the solution for certain data types. For this purpose, we propose a sparse infomax algorithm based on nonlinear Hoyer projection, leveraging both sparsity and statistical independence of latent sources. The proposed algorithm iteratively updates the unmixing matrix by infomax (for independence) and the sources by Hoyer projection (for sparsity), feeding the sparse sources back as input data for the next iteration. Consequently, sparseness propagates effectively through infomax iterations, producing sources with more desirable properties. Simulation results on both brain imaging and genetic data demonstrate that the proposed algorithm yields improved pattern recovery, particularly under low signal-to-noise ratio conditions, as well as improved sparseness compared to traditional infomax.

AB - Independent component analysis has been widely applied to brain imaging and genetic data analyses for its ability to identify interpretable latent sources. Nevertheless, leveraging source sparsity in a more granular way may further improve its ability to optimize the solution for certain data types. For this purpose, we propose a sparse infomax algorithm based on nonlinear Hoyer projection, leveraging both sparsity and statistical independence of latent sources. The proposed algorithm iteratively updates the unmixing matrix by infomax (for independence) and the sources by Hoyer projection (for sparsity), feeding the sparse sources back as input data for the next iteration. Consequently, sparseness propagates effectively through infomax iterations, producing sources with more desirable properties. Simulation results on both brain imaging and genetic data demonstrate that the proposed algorithm yields improved pattern recovery, particularly under low signal-to-noise ratio conditions, as well as improved sparseness compared to traditional infomax.

KW - Hoyer projection

KW - Imaging data

KW - Pattern recovery

KW - Snp data

KW - Sparse infomax

UR - http://www.scopus.com/inward/record.url?scp=85073889937&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073889937&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2019.8759599

DO - 10.1109/ISBI.2019.8759599

M3 - Conference contribution

C2 - 31687092

AN - SCOPUS:85073889937

T3 - Proceedings - International Symposium on Biomedical Imaging

SP - 418

EP - 421

BT - ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging

PB - IEEE Computer Society

ER -