EEG/MEG Source Reconstruction with Spatial-Temporal Two-Way Regularized Regression

Handle URI:
http://hdl.handle.net/10754/598047
Title:
EEG/MEG Source Reconstruction with Spatial-Temporal Two-Way Regularized Regression
Authors:
Tian, Tian Siva; Huang, Jianhua Z.; Shen, Haipeng; Li, Zhimin
Abstract:
In this work, we propose a spatial-temporal two-way regularized regression method for reconstructing neural source signals from EEG/MEG time course measurements. The proposed method estimates the dipole locations and amplitudes simultaneously through minimizing a single penalized least squares criterion. The novelty of our methodology is the simultaneous consideration of three desirable properties of the reconstructed source signals, that is, spatial focality, spatial smoothness, and temporal smoothness. The desirable properties are achieved by using three separate penalty functions in the penalized regression framework. Specifically, we impose a roughness penalty in the temporal domain for temporal smoothness, and a sparsity-inducing penalty and a graph Laplacian penalty in the spatial domain for spatial focality and smoothness. We develop a computational efficient multilevel block coordinate descent algorithm to implement the method. Using a simulation study with several settings of different spatial complexity and two real MEG examples, we show that the proposed method outperforms existing methods that use only a subset of the three penalty functions. © 2013 Springer Science+Business Media New York.
Citation:
Tian TS, Huang JZ, Shen H, Li Z (2013) EEG/MEG Source Reconstruction with Spatial-Temporal Two-Way Regularized Regression. Neuroinformatics 11: 477–493. Available: http://dx.doi.org/10.1007/s12021-013-9193-2.
Publisher:
Springer Nature
Journal:
Neuroinformatics
KAUST Grant Number:
KUS-CI-016-04
Issue Date:
11-Jul-2013
DOI:
10.1007/s12021-013-9193-2
PubMed ID:
23842791
Type:
Article
ISSN:
1539-2791; 1559-0089
Sponsors:
This work is supported in part by NIDA (1 RC1 DA029425-01), NSF (DMS-09-07170, DMS-10-07618, CMMI-0800575, DMS-11-06912, DMS-12-08952, and DMS-12-08786), and King Abdullah University of Science and Technology (KUS-CI-016-04).
Appears in Collections:
Publications Acknowledging KAUST Support

Full metadata record

DC FieldValue Language
dc.contributor.authorTian, Tian Sivaen
dc.contributor.authorHuang, Jianhua Z.en
dc.contributor.authorShen, Haipengen
dc.contributor.authorLi, Zhiminen
dc.date.accessioned2016-02-25T13:11:39Zen
dc.date.available2016-02-25T13:11:39Zen
dc.date.issued2013-07-11en
dc.identifier.citationTian TS, Huang JZ, Shen H, Li Z (2013) EEG/MEG Source Reconstruction with Spatial-Temporal Two-Way Regularized Regression. Neuroinformatics 11: 477–493. Available: http://dx.doi.org/10.1007/s12021-013-9193-2.en
dc.identifier.issn1539-2791en
dc.identifier.issn1559-0089en
dc.identifier.pmid23842791en
dc.identifier.doi10.1007/s12021-013-9193-2en
dc.identifier.urihttp://hdl.handle.net/10754/598047en
dc.description.abstractIn this work, we propose a spatial-temporal two-way regularized regression method for reconstructing neural source signals from EEG/MEG time course measurements. The proposed method estimates the dipole locations and amplitudes simultaneously through minimizing a single penalized least squares criterion. The novelty of our methodology is the simultaneous consideration of three desirable properties of the reconstructed source signals, that is, spatial focality, spatial smoothness, and temporal smoothness. The desirable properties are achieved by using three separate penalty functions in the penalized regression framework. Specifically, we impose a roughness penalty in the temporal domain for temporal smoothness, and a sparsity-inducing penalty and a graph Laplacian penalty in the spatial domain for spatial focality and smoothness. We develop a computational efficient multilevel block coordinate descent algorithm to implement the method. Using a simulation study with several settings of different spatial complexity and two real MEG examples, we show that the proposed method outperforms existing methods that use only a subset of the three penalty functions. © 2013 Springer Science+Business Media New York.en
dc.description.sponsorshipThis work is supported in part by NIDA (1 RC1 DA029425-01), NSF (DMS-09-07170, DMS-10-07618, CMMI-0800575, DMS-11-06912, DMS-12-08952, and DMS-12-08786), and King Abdullah University of Science and Technology (KUS-CI-016-04).en
dc.publisherSpringer Natureen
dc.subjectCoordinate descenten
dc.subjectGraph Laplacianen
dc.subjectInverse problemen
dc.subjectMEGen
dc.subjectRoughness penalizationen
dc.subjectSparsityen
dc.titleEEG/MEG Source Reconstruction with Spatial-Temporal Two-Way Regularized Regressionen
dc.typeArticleen
dc.identifier.journalNeuroinformaticsen
dc.contributor.institutionUniversity of Houston, Houston, United Statesen
dc.contributor.institutionTexas A and M University, College Station, United Statesen
dc.contributor.institutionThe University of North Carolina at Chapel Hill, Chapel Hill, United Statesen
dc.contributor.institutionMedical College of Wisconsin, Milwaukee, United Statesen
kaust.grant.numberKUS-CI-016-04en

Related articles on PubMed

All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.