Staff View: Estimating brain connectivity using copula Gaussian graphical models

Estimating brain connectivity using copula Gaussian graphical models

Electroencephalogram (EEG) has been widely used to study cortical connectivity during acquisition of motor skills. Previous studies using graphical models to estimate sparse brain networks focused on time-domain dependency. This paper introduces graphical models in the spectral domain to characteriz...

Full description

Saved in:

Bibliographic Details
Main Authors:	Gao, Xu, Shen, Weining, Ting, Chee Ming, Cramer, Steven C., Srinivasan, Ramesh, Ombao, Hernando
Format:	Conference or Workshop Item
Published:	2019
Subjects:	QH301 Biology TA Engineering (General). Civil engineering (General)
Online Access:	http://eprints.utm.my/id/eprint/89563/ http://dx.doi.org/10.1109/ISBI.2019.8759538
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my.utm.89563
record_format	eprints
spelling	my.utm.895632021-02-22T06:10:24Z http://eprints.utm.my/id/eprint/89563/ Estimating brain connectivity using copula Gaussian graphical models Gao, Xu Shen, Weining Ting, Chee Ming Cramer, Steven C. Srinivasan, Ramesh Ombao, Hernando QH301 Biology TA Engineering (General). Civil engineering (General) Electroencephalogram (EEG) has been widely used to study cortical connectivity during acquisition of motor skills. Previous studies using graphical models to estimate sparse brain networks focused on time-domain dependency. This paper introduces graphical models in the spectral domain to characterize dependence in oscillatory activity between EEG channels. We first apply a transformation based on a copula Gaussian graphical model to deal with non-Gaussianity in the data. To obtain a simple and robust representation of brain connectivity that explains most variation in the data, we propose a framework based on maximizing penalized likelihood with Lasso regularization utilizing the cross-spectral density matrix to search for a sparse precision matrix. To solve the optimization problem, we developed modified versions of graphical Lasso, Ledoit-Wolf (LW) and the majorize-minimize sparse covariance estimation (SPCOV) algorithms. Simulations show benefits of the proposed algorithms in terms of robustness and accurate estimation under non-Gaussianity and different structures of high-dimensional sparse networks. On EEG data of a motor skill task, the modified graphical Lasso and LW algorithms reveal sparse connectivity pattern among cortices in consistency with previous findings. In addition, our results suggest regions over different frequency bands yield distinct impacts on motor skill learning. 2019-04 Conference or Workshop Item PeerReviewed Gao, Xu and Shen, Weining and Ting, Chee Ming and Cramer, Steven C. and Srinivasan, Ramesh and Ombao, Hernando (2019) Estimating brain connectivity using copula Gaussian graphical models. In: 16th IEEE International Symposium on Biomedical Imaging, ISBI 2019, 8 April 2019 through 11 April 2019, Hilton Molino Stucky - Venice, Venice, Italy. http://dx.doi.org/10.1109/ISBI.2019.8759538
institution	Universiti Teknologi Malaysia
building	UTM Library
collection	Institutional Repository
continent	Asia
country	Malaysia
content_provider	Universiti Teknologi Malaysia
content_source	UTM Institutional Repository
url_provider	http://eprints.utm.my/
topic	QH301 Biology TA Engineering (General). Civil engineering (General)
spellingShingle	QH301 Biology TA Engineering (General). Civil engineering (General) Gao, Xu Shen, Weining Ting, Chee Ming Cramer, Steven C. Srinivasan, Ramesh Ombao, Hernando Estimating brain connectivity using copula Gaussian graphical models
description	Electroencephalogram (EEG) has been widely used to study cortical connectivity during acquisition of motor skills. Previous studies using graphical models to estimate sparse brain networks focused on time-domain dependency. This paper introduces graphical models in the spectral domain to characterize dependence in oscillatory activity between EEG channels. We first apply a transformation based on a copula Gaussian graphical model to deal with non-Gaussianity in the data. To obtain a simple and robust representation of brain connectivity that explains most variation in the data, we propose a framework based on maximizing penalized likelihood with Lasso regularization utilizing the cross-spectral density matrix to search for a sparse precision matrix. To solve the optimization problem, we developed modified versions of graphical Lasso, Ledoit-Wolf (LW) and the majorize-minimize sparse covariance estimation (SPCOV) algorithms. Simulations show benefits of the proposed algorithms in terms of robustness and accurate estimation under non-Gaussianity and different structures of high-dimensional sparse networks. On EEG data of a motor skill task, the modified graphical Lasso and LW algorithms reveal sparse connectivity pattern among cortices in consistency with previous findings. In addition, our results suggest regions over different frequency bands yield distinct impacts on motor skill learning.
format	Conference or Workshop Item
author	Gao, Xu Shen, Weining Ting, Chee Ming Cramer, Steven C. Srinivasan, Ramesh Ombao, Hernando
author_facet	Gao, Xu Shen, Weining Ting, Chee Ming Cramer, Steven C. Srinivasan, Ramesh Ombao, Hernando
author_sort	Gao, Xu
title	Estimating brain connectivity using copula Gaussian graphical models
title_short	Estimating brain connectivity using copula Gaussian graphical models
title_full	Estimating brain connectivity using copula Gaussian graphical models
title_fullStr	Estimating brain connectivity using copula Gaussian graphical models
title_full_unstemmed	Estimating brain connectivity using copula Gaussian graphical models
title_sort	estimating brain connectivity using copula gaussian graphical models
publishDate	2019
url	http://eprints.utm.my/id/eprint/89563/ http://dx.doi.org/10.1109/ISBI.2019.8759538
_version_	1693725942173663232
score	13.211869

Estimating brain connectivity using copula Gaussian graphical models

Similar Items