MEG connectivity and power detections with minimum norm estimates require different regularization parameters

Research output: Contribution to journalArticleScientificpeer-review

Standard

MEG connectivity and power detections with minimum norm estimates require different regularization parameters. / Hincapié, Ana Sofía; Kujala, Jan; Mattout, Jérémie; Daligault, Sebastien; Delpuech, Claude; Mery, Domingo; Cosmelli, Diego; Jerbi, Karim.

In: COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE, Vol. 2016, 3979547, 2016, p. 1-11.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

APA

Vancouver

Author

Hincapié, Ana Sofía ; Kujala, Jan ; Mattout, Jérémie ; Daligault, Sebastien ; Delpuech, Claude ; Mery, Domingo ; Cosmelli, Diego ; Jerbi, Karim. / MEG connectivity and power detections with minimum norm estimates require different regularization parameters. In: COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE. 2016 ; Vol. 2016. pp. 1-11.

Bibtex - Download

@article{07794d7e722e4d28b8e9008c382f772b,
title = "MEG connectivity and power detections with minimum norm estimates require different regularization parameters",
abstract = "Minimum Norm Estimation (MNE) is an inverse solution method widely used to reconstruct the source time series that underlie magnetoencephalography (MEG) data. MNE addresses the ill-posed nature of MEG source estimation through regularization (e.g., Tikhonov regularization). Selecting the best regularization parameter is a critical step. Generally, once set, it is common practice to keep the same coefficient throughout a study. However, it is yet to be known whether the optimal lambda for spectral power analysis of MEG source data coincides with the optimal regularization for source-level oscillatory coupling analysis. We addressed this question via extensive Monte-Carlo simulations of MEG data, where we generated 21,600 configurations of pairs of coupled sources with varying sizes, signal-to-noise ratio (SNR), and coupling strengths. Then, we searched for the Tikhonov regularization coefficients (lambda) that maximize detection performance for (a) power and (b) coherence. For coherence, the optimal lambda was two orders of magnitude smaller than the best lambda for power. Moreover, we found that the spatial extent of the interacting sources and SNR, but not the extent of coupling, were the main parameters affecting the best choice for lambda. Our findings suggest using less regularization when measuring oscillatory coupling compared to power estimation.",
author = "Hincapi{\'e}, {Ana Sof{\'i}a} and Jan Kujala and J{\'e}r{\'e}mie Mattout and Sebastien Daligault and Claude Delpuech and Domingo Mery and Diego Cosmelli and Karim Jerbi",
year = "2016",
doi = "10.1155/2016/3979547",
language = "English",
volume = "2016",
pages = "1--11",
journal = "COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE",
issn = "1687-5265",
publisher = "Hindawi Publishing Corporation",

}

RIS - Download

TY - JOUR

T1 - MEG connectivity and power detections with minimum norm estimates require different regularization parameters

AU - Hincapié, Ana Sofía

AU - Kujala, Jan

AU - Mattout, Jérémie

AU - Daligault, Sebastien

AU - Delpuech, Claude

AU - Mery, Domingo

AU - Cosmelli, Diego

AU - Jerbi, Karim

PY - 2016

Y1 - 2016

N2 - Minimum Norm Estimation (MNE) is an inverse solution method widely used to reconstruct the source time series that underlie magnetoencephalography (MEG) data. MNE addresses the ill-posed nature of MEG source estimation through regularization (e.g., Tikhonov regularization). Selecting the best regularization parameter is a critical step. Generally, once set, it is common practice to keep the same coefficient throughout a study. However, it is yet to be known whether the optimal lambda for spectral power analysis of MEG source data coincides with the optimal regularization for source-level oscillatory coupling analysis. We addressed this question via extensive Monte-Carlo simulations of MEG data, where we generated 21,600 configurations of pairs of coupled sources with varying sizes, signal-to-noise ratio (SNR), and coupling strengths. Then, we searched for the Tikhonov regularization coefficients (lambda) that maximize detection performance for (a) power and (b) coherence. For coherence, the optimal lambda was two orders of magnitude smaller than the best lambda for power. Moreover, we found that the spatial extent of the interacting sources and SNR, but not the extent of coupling, were the main parameters affecting the best choice for lambda. Our findings suggest using less regularization when measuring oscillatory coupling compared to power estimation.

AB - Minimum Norm Estimation (MNE) is an inverse solution method widely used to reconstruct the source time series that underlie magnetoencephalography (MEG) data. MNE addresses the ill-posed nature of MEG source estimation through regularization (e.g., Tikhonov regularization). Selecting the best regularization parameter is a critical step. Generally, once set, it is common practice to keep the same coefficient throughout a study. However, it is yet to be known whether the optimal lambda for spectral power analysis of MEG source data coincides with the optimal regularization for source-level oscillatory coupling analysis. We addressed this question via extensive Monte-Carlo simulations of MEG data, where we generated 21,600 configurations of pairs of coupled sources with varying sizes, signal-to-noise ratio (SNR), and coupling strengths. Then, we searched for the Tikhonov regularization coefficients (lambda) that maximize detection performance for (a) power and (b) coherence. For coherence, the optimal lambda was two orders of magnitude smaller than the best lambda for power. Moreover, we found that the spatial extent of the interacting sources and SNR, but not the extent of coupling, were the main parameters affecting the best choice for lambda. Our findings suggest using less regularization when measuring oscillatory coupling compared to power estimation.

UR - http://www.scopus.com/inward/record.url?scp=84964811351&partnerID=8YFLogxK

U2 - 10.1155/2016/3979547

DO - 10.1155/2016/3979547

M3 - Article

VL - 2016

SP - 1

EP - 11

JO - COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE

JF - COMPUTATIONAL INTELLIGENCE AND NEUROSCIENCE

SN - 1687-5265

M1 - 3979547

ER -

ID: 3264985