Comparison of beamformer implementations for MEG source localization

Amit Jaiswal,Jukka Nenonen,Matti Stenroos,Alexandre Gramfort,Sarang S Dalal,Britta U Westner,Vladimir Litvak,John C Mosher,Jan-Mathijs Schoffelen,Caroline Witton,Robert Oostenveld,Lauri Parkkonen

doi:10.1016/j.neuroimage.2020.116797

Abstract

Beamformers are applied for estimating spatiotemporal characteristics of neuronal sources underlying measured MEG/EEG signals. Several MEG analysis toolboxes include an implementation of a linearly constrained minimum-variance (LCMV) beamformer. However, differences in implementations and in their results complicate the selection and application of beamformers and may hinder their wider adoption in research and clinical use. Additionally, combinations of different MEG sensor types (such as magnetometers and planar gradiometers) and application of preprocessing methods for interference suppression, such as signal space separation (SSS), can affect the results in different ways for different implementations. So far, a systematic evaluation of the different implementations has not been performed. Here, we compared the localization performance of the LCMV beamformer pipelines in four widely used open-source toolboxes (MNE-Python, FieldTrip, DAiSS (SPM12), and Brainstorm) using datasets both with and without SSS interference suppression.We analyzed MEG data that were i) simulated, ii) recorded from a static and moving phantom, and iii) recorded from a healthy volunteer receiving auditory, visual, and somatosensory stimulation. We also investigated the effects of SSS and the combination of the magnetometer and gradiometer signals. We quantified how localization error and point-spread volume vary with the signal-to-noise ratio (SNR) in all four toolboxes.When applied carefully to MEG data with a typical SNR (3–15 dB), all four toolboxes localized the sources reliably; however, they differed in their sensitivity to preprocessing parameters. As expected, localizations were highly unreliable at very low SNR, but we found high localization error also at very high SNRs for the first three toolboxes while Brainstorm showed greater robustness but with lower spatial resolution. We also found that the SNR improvement offered by SSS led to more accurate localization.

Highlights

MEG and EEG source imaging aims to identify the spatiotemporal characteristics of neural source currents based on the recorded signals, electromagnetic forward models and physiologically motivated assumptions about the source distribution
Localizations were highly unreliable at very low signal-to-noise ratio (SNR), but we found high localization error at very high SNRs for the first three toolboxes while Brainstorm showed greater robustness but with lower spatial resolution
We found that the SNR improvement offered by signal space separation (SSS) led to more accurate localization

Summary

Introduction

MEG (magnetoencephalography) and EEG (electroencephalography) source imaging aims to identify the spatiotemporal characteristics of neural source currents based on the recorded signals, electromagnetic forward models and physiologically motivated assumptions about the source distribution. One well-known method for estimating a small number of focal sources is to model each of them as a current dipole with fixed location and fixed or changing orientation. The locations (optionally orientations) and time courses of the dipoles are collectively estimated (Mosher et al, 1992; H€am€al€ainen et al, 1993). Such equivalent dipole models have been widely applied in basic research Examples of linear methods for distributed source estimation are LORETA (low-resolution brain electromagnetic tomography; Pascual-Marqui et al, 1994) and MNE (minimum-norm estimation; H€am€al€ainen and Ilmoniemi, 1994). Various non-linear distributed inverse methods have been proposed (Wipf et al, 2010; Gramfort et al, 2013b)

Methods

Results

Discussion

Conclusion