An Iterative Implementation of the Signal Space Separation Method for Magnetoencephalography Systems with Low Channel Counts

Abstract

The signal space separation (SSS) method is routinely employed in the analysis of multichannel magnetic field recordings (such as magnetoencephalography (MEG) data). In the SSS method, signal vectors are posed as a multipole expansion of the magnetic field, allowing contributions from sources internal and external to a sensor array to be separated via computation of the pseudo-inverse of a matrix of the basis vectors. Although powerful, the standard implementation of the SSS method on MEG systems based on optically pumped magnetometers (OPMs) is unstable due to the approximate parity of the required number of dimensions of the SSS basis and the number of channels in the data. Here we exploit the hierarchical nature of the multipole expansion to perform a stable, iterative implementation of the SSS method. We describe the method and investigate its performance via a simulation study on a 192-channel OPM-MEG helmet. We assess performance for different levels of truncation of the SSS basis and a varying number of iterations. Results show that the iterative method provides stable performance, with a clear separation of internal and external sources.

Adaptive multipole models of OPM data

“…We also could have excluded harmonics until the condition number reached an acceptable threshold (as can done in the SPM implementation for cryogenic MEG systems). We could have also selected the harmonics to be modelled using simulations that maximized SNR (Wang et al, 2023) or constructed the SSS design matrix in an iterative manner (Holmes et al, 2023). However, these approaches would need to be adapted for each and every system and our focus here is to provide a method that has more deterministic performance across a wide variety of OPM systems, with minimal custom modification.…”

“…Software based correction of motion artefacts and environmental interference is a key (and currently unresolved) issue in the development of wearable OPM systems (Brookes et al, 2021; Ding et al, 2023; Hill et al, 2022; Holmes et al, 2023; Mellor et al, 2021; Nardelli et al, 2020; Rea et al, 2021; Seymour et al, 2021, 2022; Tierney et al, 2021; Wang et al, 2023). Thankfully, multipole models of magnetic field are uniquely suited for suppressing the interference encountered in this scenario (Tierney et al, 2022).…”

On-scalp magnetoencephalography can detect mesial temporal lobe epileptiform discharges

“…It thus opens the door for future on-scalp or close-to-scalp OPM-MEG recordings in adverse recording situations, such as SEEG to overcome its limited spatial brain sampling or implanted neuromodulation devices (e.g., responsive neurostimulation). Specific OPM-MEG denoising methods such as signal space separation 15 could increase the IED detection rate of OPM-MEG in patients with implanted intracranial materials.…”

Pushing the boundaries of MEG based on optically pumped magnetometers towards early human life

“…This can be addressed by keeping the ambient field as low as possible, which requires magnetic shielding and the use of field-nulling coils, as applied in our study; however, a small magnetic field always remains. Further improvements in interference suppression techniques [40,41] will likely increase the SNR of OPM data which may exceed that of cryogenic MEG data. Indeed, in our comparison between OPM-MEG and cryogenic MEG, we found similar SNRs despite the significantly higher signal amplitude with OPM-MEG.…”

“…Further improvements in interference suppression techniques (40,41) will likely increase the SNR of OPM data which may exceed that of cryogenic MEG data. Indeed, in our comparison between OPM-MEG and cryogenic MEG, we found similar SNRs despite the significantly higher signal amplitude with OPM-MEG.…”