Expectation–maximization algorithm with a nonlinear kalman smoother for MEG/EEG connectivity estimation

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Standard

Expectation–maximization algorithm with a nonlinear kalman smoother for MEG/EEG connectivity estimation. / Subramaniyam, Narayan Puthanmadam; Tronarp, Filip; Särkkä, Simo; Parkkonen, Lauri.

EMBEC and NBC 2017 - Joint Conference of the European Medical and Biological Engineering Conference EMBEC 2017 and the Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2017. 2017. p. 763-766 (IFMBE Proceedings; Vol. 65).

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Harvard

Subramaniyam, NP, Tronarp, F, Särkkä, S & Parkkonen, L 2017, Expectation–maximization algorithm with a nonlinear kalman smoother for MEG/EEG connectivity estimation. in EMBEC and NBC 2017 - Joint Conference of the European Medical and Biological Engineering Conference EMBEC 2017 and the Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2017. IFMBE Proceedings, vol. 65, pp. 763-766, Joint Conference of the European Medical and Biological Engineering Conference, EMBEC 2017 and Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2107, Tampere, Finland, 11/06/2017. https://doi.org/10.1007/978-981-10-5122-7_191

APA

Subramaniyam, N. P., Tronarp, F., Särkkä, S., & Parkkonen, L. (2017). Expectation–maximization algorithm with a nonlinear kalman smoother for MEG/EEG connectivity estimation. In EMBEC and NBC 2017 - Joint Conference of the European Medical and Biological Engineering Conference EMBEC 2017 and the Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2017 (pp. 763-766). (IFMBE Proceedings; Vol. 65). https://doi.org/10.1007/978-981-10-5122-7_191

Vancouver

Subramaniyam NP, Tronarp F, Särkkä S, Parkkonen L. Expectation–maximization algorithm with a nonlinear kalman smoother for MEG/EEG connectivity estimation. In EMBEC and NBC 2017 - Joint Conference of the European Medical and Biological Engineering Conference EMBEC 2017 and the Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2017. 2017. p. 763-766. (IFMBE Proceedings). https://doi.org/10.1007/978-981-10-5122-7_191

Author

Subramaniyam, Narayan Puthanmadam ; Tronarp, Filip ; Särkkä, Simo ; Parkkonen, Lauri. / Expectation–maximization algorithm with a nonlinear kalman smoother for MEG/EEG connectivity estimation. EMBEC and NBC 2017 - Joint Conference of the European Medical and Biological Engineering Conference EMBEC 2017 and the Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2017. 2017. pp. 763-766 (IFMBE Proceedings).

Bibtex - Download

@inproceedings{d51d75167e1144f1ba5f011c25128d80,
title = "Expectation–maximization algorithm with a nonlinear kalman smoother for MEG/EEG connectivity estimation",
abstract = "Current techniques to determine functional or effective connectivity from magnetoencephalography (MEG) and electroencephalography (EEG) signals typically involve two sequential steps: 1) estimation of the source current distribution from the sensor data, for example, by minimum-norm estimation or beamforming, and 2) fitting a multivariate autoregressive (MVAR) model to estimate the AR coefficients, which reflect the interaction between the sources. Here, we introduce a combination of the expectation–maximization (EM) algorithm and a nonlinear Kalman smoother to perform joint estimation of both source and connectivity (linear and nonlinear) parameters from MEG/EEG signals. Based on simulations, we show that the proposed approach estimates both the source signals and AR coefficients in linear models significantly better than the traditional two-step approach when the signal-to-noise ratio (SNR) is low (≤1) and gives comparable results at higher SNRs (>1). Additionally, we show that nonlinear interaction parameters can be reliably estimated from MEG/EEG signals at low SNRs using the EM algorithm with sigma-point Kalman smoother.",
keywords = "EEG, Expectation-maximization, Functional connectivity, MEG, Nonlinear Kalman smoother",
author = "Subramaniyam, {Narayan Puthanmadam} and Filip Tronarp and Simo S{\"a}rkk{\"a} and Lauri Parkkonen",
year = "2017",
doi = "10.1007/978-981-10-5122-7_191",
language = "English",
isbn = "9789811051210",
series = "IFMBE Proceedings",
publisher = "Springer-Verlag",
pages = "763--766",
booktitle = "EMBEC and NBC 2017 - Joint Conference of the European Medical and Biological Engineering Conference EMBEC 2017 and the Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2017",

}

RIS - Download

TY - GEN

T1 - Expectation–maximization algorithm with a nonlinear kalman smoother for MEG/EEG connectivity estimation

AU - Subramaniyam, Narayan Puthanmadam

AU - Tronarp, Filip

AU - Särkkä, Simo

AU - Parkkonen, Lauri

PY - 2017

Y1 - 2017

N2 - Current techniques to determine functional or effective connectivity from magnetoencephalography (MEG) and electroencephalography (EEG) signals typically involve two sequential steps: 1) estimation of the source current distribution from the sensor data, for example, by minimum-norm estimation or beamforming, and 2) fitting a multivariate autoregressive (MVAR) model to estimate the AR coefficients, which reflect the interaction between the sources. Here, we introduce a combination of the expectation–maximization (EM) algorithm and a nonlinear Kalman smoother to perform joint estimation of both source and connectivity (linear and nonlinear) parameters from MEG/EEG signals. Based on simulations, we show that the proposed approach estimates both the source signals and AR coefficients in linear models significantly better than the traditional two-step approach when the signal-to-noise ratio (SNR) is low (≤1) and gives comparable results at higher SNRs (>1). Additionally, we show that nonlinear interaction parameters can be reliably estimated from MEG/EEG signals at low SNRs using the EM algorithm with sigma-point Kalman smoother.

AB - Current techniques to determine functional or effective connectivity from magnetoencephalography (MEG) and electroencephalography (EEG) signals typically involve two sequential steps: 1) estimation of the source current distribution from the sensor data, for example, by minimum-norm estimation or beamforming, and 2) fitting a multivariate autoregressive (MVAR) model to estimate the AR coefficients, which reflect the interaction between the sources. Here, we introduce a combination of the expectation–maximization (EM) algorithm and a nonlinear Kalman smoother to perform joint estimation of both source and connectivity (linear and nonlinear) parameters from MEG/EEG signals. Based on simulations, we show that the proposed approach estimates both the source signals and AR coefficients in linear models significantly better than the traditional two-step approach when the signal-to-noise ratio (SNR) is low (≤1) and gives comparable results at higher SNRs (>1). Additionally, we show that nonlinear interaction parameters can be reliably estimated from MEG/EEG signals at low SNRs using the EM algorithm with sigma-point Kalman smoother.

KW - EEG

KW - Expectation-maximization

KW - Functional connectivity

KW - MEG

KW - Nonlinear Kalman smoother

UR - http://www.scopus.com/inward/record.url?scp=85021750408&partnerID=8YFLogxK

U2 - 10.1007/978-981-10-5122-7_191

DO - 10.1007/978-981-10-5122-7_191

M3 - Conference contribution

SN - 9789811051210

T3 - IFMBE Proceedings

SP - 763

EP - 766

BT - EMBEC and NBC 2017 - Joint Conference of the European Medical and Biological Engineering Conference EMBEC 2017 and the Nordic-Baltic Conference on Biomedical Engineering and Medical Physics, NBC 2017

ER -

ID: 14299881