Estimating Dynamic Connectivity States in fMRI Using Regime-Switching Factor Models

Handle URI:
http://hdl.handle.net/10754/626407
Title:
Estimating Dynamic Connectivity States in fMRI Using Regime-Switching Factor Models
Authors:
Ting, Chee-Ming; Ombao, Hernando; Samdin, S. Balqis; Salleh, Sh-Hussain
Abstract:
We consider the challenges in estimating state-related changes in brain connectivity networks with a large number of nodes. Existing studies use sliding-window analysis or time-varying coefficient models which are unable to capture both smooth and abrupt changes simultaneously, and rely on ad-hoc approaches to the high-dimensional estimation. To overcome these limitations, we propose a Markov-switching dynamic factor model which allows the dynamic connectivity states in functional magnetic resonance imaging (fMRI) data to be driven by lower-dimensional latent factors. We specify a regime-switching vector autoregressive (SVAR) factor process to quantity the time-varying directed connectivity. The model enables a reliable, data-adaptive estimation of change-points of connectivity regimes and the massive dependencies associated with each regime. We develop a three-step estimation procedure: 1) extracting the factors using principal component analysis, 2) identifying connectivity regimes in a low-dimensional subspace based on the factor-based SVAR model, 3) constructing high-dimensional state connectivity metrics based on the subspace estimates. Simulation results show that our estimator outperforms K-means clustering of time-windowed coefficients, providing more accurate estimate of time-evolving connectivity. It achieves percentage of reduction in mean squared error by 60% when the network dimension is comparable to the sample size. When applied to resting-state fMRI data, our method successfully identifies modular organization in resting-state networks in consistency with other studies. It further reveals changes in brain states with variations across subjects and distinct large-scale directed connectivity patterns across states.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Citation:
Ting C-M, Ombao H, Samdin SB, Salleh S-H (2017) Estimating Dynamic Connectivity States in fMRI Using Regime-Switching Factor Models. IEEE Transactions on Medical Imaging: 1–1. Available: http://dx.doi.org/10.1109/TMI.2017.2780185.
Publisher:
Institute of Electrical and Electronics Engineers (IEEE)
Journal:
IEEE Transactions on Medical Imaging
Issue Date:
6-Dec-2017
DOI:
10.1109/TMI.2017.2780185
Type:
Article
ISSN:
0278-0062; 1558-254X
Appears in Collections:
Articles; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorTing, Chee-Mingen
dc.contributor.authorOmbao, Hernandoen
dc.contributor.authorSamdin, S. Balqisen
dc.contributor.authorSalleh, Sh-Hussainen
dc.date.accessioned2017-12-21T13:57:04Z-
dc.date.available2017-12-21T13:57:04Z-
dc.date.issued2017-12-06en
dc.identifier.citationTing C-M, Ombao H, Samdin SB, Salleh S-H (2017) Estimating Dynamic Connectivity States in fMRI Using Regime-Switching Factor Models. IEEE Transactions on Medical Imaging: 1–1. Available: http://dx.doi.org/10.1109/TMI.2017.2780185.en
dc.identifier.issn0278-0062en
dc.identifier.issn1558-254Xen
dc.identifier.doi10.1109/TMI.2017.2780185en
dc.identifier.urihttp://hdl.handle.net/10754/626407-
dc.description.abstractWe consider the challenges in estimating state-related changes in brain connectivity networks with a large number of nodes. Existing studies use sliding-window analysis or time-varying coefficient models which are unable to capture both smooth and abrupt changes simultaneously, and rely on ad-hoc approaches to the high-dimensional estimation. To overcome these limitations, we propose a Markov-switching dynamic factor model which allows the dynamic connectivity states in functional magnetic resonance imaging (fMRI) data to be driven by lower-dimensional latent factors. We specify a regime-switching vector autoregressive (SVAR) factor process to quantity the time-varying directed connectivity. The model enables a reliable, data-adaptive estimation of change-points of connectivity regimes and the massive dependencies associated with each regime. We develop a three-step estimation procedure: 1) extracting the factors using principal component analysis, 2) identifying connectivity regimes in a low-dimensional subspace based on the factor-based SVAR model, 3) constructing high-dimensional state connectivity metrics based on the subspace estimates. Simulation results show that our estimator outperforms K-means clustering of time-windowed coefficients, providing more accurate estimate of time-evolving connectivity. It achieves percentage of reduction in mean squared error by 60% when the network dimension is comparable to the sample size. When applied to resting-state fMRI data, our method successfully identifies modular organization in resting-state networks in consistency with other studies. It further reveals changes in brain states with variations across subjects and distinct large-scale directed connectivity patterns across states.en
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en
dc.rights(c) 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.en
dc.subjectBrain modelingen
dc.subjectCovariance matricesen
dc.subjectdynamic brain connectivityen
dc.subjectEstimationen
dc.subjectfactor analysisen
dc.subjectfMRIen
dc.subjectHidden Markov modelsen
dc.subjectlarge VAR modelsen
dc.subjectLoad modelingen
dc.subjectReactive poweren
dc.subjectRegime-switching modelsen
dc.titleEstimating Dynamic Connectivity States in fMRI Using Regime-Switching Factor Modelsen
dc.typeArticleen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.identifier.journalIEEE Transactions on Medical Imagingen
dc.eprint.versionPost-printen
dc.contributor.institutionCenter for Biomedical Engineering (CBE), Universiti Teknologi Malaysia (UTM), 81310 Skudai, Johor, Malaysia.en
kaust.authorTing, Chee-Mingen
kaust.authorOmbao, Hernandoen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.