Pragmatic spatial sampling for wearable MEG arrays

Abstract

Several new technologies have recently emerged promising new MEG systems in which the sensors can be placed close to the scalp. One such technology, Optically Pumped Magnetometry MEG (OP-MEG) allows for a scalp mounted flexible system that provides field measurements within mm of the scalp surface. A question that arises in developing on-scalp systems, such as OP-MEG scanners, is: how many sensors are necessary to achieve adequate performance/spatial discrimination? There are many factors to consider in answering this question such as the signal to noise ratio (SNR), the locations and depths of the sources of interest, the density of spatial sampling, sensor gain errors (due to interference, subject movement, cross-talk, etc.) and, of course, the desired spatial discrimination. In this paper, we provide simulations which show the impact these factors have on designing sensor arrays for wearable MEG. While OP-MEG has the potential to provide high information content at dense spatial samplings, we find that adequate spatial discrimination of sources (<1cm) can be achieved with relatively few sensors (<100) at coarse spatial samplings (~30mm) at high SNR. Comparable discrimination for traditional cryogenic systems require far more channels by these same metrics. Finally we show that sensor gain errors have the greatest impact on discrimination between deep sources at high SNR.

Multi-channel whole-head OPM-MEG: Helmet design and a comparison with a conventional system

“…With this in mind, perhaps the most surprising result of this study is that source localisation is comparable between OPM and cryogenic systems, despite the fact that the cryogenic system has more than 5 times more sensors. However, this appears to support the theoretical findings by Tierney et al., 2019b . It is tempting to speculate that whilst, here, we have shown our OPM system to be “as good” as the current state of the art, as OPM systems inevitably gain more sensors, it is likely that they will significantly overtake cryogenic instruments.…”

“…Sensor array coverage (i.e. where to place OPMs to cover all possible cortical locations) and the number of OPMs required to gain parity of performance with cryogenic systems remain open questions ( Iivanainen et al, 2019a ; Tierney et al, 2019b ) and there are, to date, few studies comparing OPM and SQUID measurements ( Borna, 2020 ; Boto et al, 2017 ; Iivanainen et al, 2019c ). Such comparisons are critical if the MEG community is to gain confidence and adopt OPM technology.…”

Optically pumped magnetoencephalography in epilepsy

“…The advantages of conventional MEG in this case would have been increased channel number and whole‐head coverage. We are currently working toward a system using 50 OPM devices where we should expect to see comparable whole‐brain source‐level discrimination to a cryogenic system with approximately 250 sensors. Further development of the technique will include simultaneous 30‐channel OP‐MEG and EEG recording, which has already been demonstrated in healthy subjects and prolonged EEG/OP‐MEG recording over several hours with video monitoring.…”

Modelling optically pumped magnetometer interference as a mean (magnetic) field

“…For instance, inherent to DSSP is the assumption that the rank of the data is much greater than the rank of the lead fields. This is clearly the case in cryogenic MEG systems that may have 300 sensors and a lead field rank in the range of 50-100 (Iivanainen et al, 2020; Nenonen, Taulu, Kajola, & Ahonen, 2007; Tierney, Mellor, et al, 2019). In OPM systems this may not be the case and the rank of the lead fields may be comparable to the rank of the data because typical systems currently operate with fewer than 50 sensors (Hill et al, 2020).…”