Abstracts

Rationale: Despite its recognized clinical-added value for the non-invasive localization of the irritative zone in patients with refractory focal epilepsy¹, the use of magnetoencephalography (MEG) is still rather limited worldwide. This is mainly due to the sensor technology, which is at the basis of the main limitations of MEG. Current MEG systems house hundreds of cryogenic sensors in a rigid, one-size-fits-all helmet requiring a thermally insulating gap between the scalp and the sensors. This gap leads to substantial MEG signal attenuation (magnetic fields decrease with the square of the distance) due to a brain-to-sensor distance of about 2-5 cm in adults, that is even larger in children. Optically-pumped magnetometers (OPMs) are novel cryogen-free magnetic field sensors that can be placed directly on the scalp.² On-scalp MEG based on OPMs (OPM-MEG) can therefore be adapted to any head shape/size and record human brain activity with an expected increased signal-to-noise ratio (SNR) and spatial resolution.² A direct comparison of OPM-MEG with cryogenic MEG is still missing in the field of epilepsy. Here, we study OPM-MEG in the routine clinical use by comparing interictal epileptiform discharges (IEDs) recorded with OPMs fixed on EEG-like caps and cryogenic MEG in five epileptic children.

Methods: Five children with focal epilepsy (4F; age range, 5-11y; 3 self-limited genetic and 2 refractory focal epilepsy) underwent cryogenic and OPM-MEG in the siting position while watching a movie. OPM-MEG was recorded with 32 OPMs fixed on 3D printed plastic sensor mounts sewn on flexible EEG caps and placed around the presumed epileptogenic zone (Fig. 1). Cryogenic MEG was restricted to its 102 magnetometers. IEDs were visually identified in MEG signals. A spike-wave index (SWI)³ was computed. IED peak amplitude and SNR were compared. The neural source of averaged IEDs were localized using distributed source modeling.

Results: IEDs were found in all children (3 left centro-temporal (CT), 1 right CT, 1 left frontal; SWI, 2-89%) with higher amplitude with OPM-MEG in all patients (2.3-4.8 times higher) due to reduced brain-to-sensor distance with OPMs (24.5-33.2mm) compared with cryogenic magnetometers (48.7-67.1mm). SNR was higher with OPM-MEG in four patients (27-60% higher) and similar in the other due to high amplitude head movement artefacts. Reconstructed neural sources of IEDs between both types of MEG were distant from 4.2mm to 15.6mm (3 in the range of cryogenic MEG spatial resolution (about 5mm), 2 higher; Fig. 2).

Conclusions: On-scalp OPM-MEG has all the potential to replace cryogenic MEG as a method of reference in the assessment of focal epilepsy, allowing routine ictal MEG and video-MEG recordings at reduced cost.

Funding: Please list any funding that was received in support of this abstract.: OF is supported by a research grant from the FRIA. PC is supported by a research grant from the Fonds Erasme. XDT is Postdoctorate Clinical Master Specialist at the FRS-FNRS. This study has been supported by the Fonds Erasme and the FRS-FNRS. The OPM- and SQUID-MEG projects at the CUB Hôpital Erasme are financially supported by the Fonds Erasme.