Back

Detection of mesial temporal lobe epilepsy with OP-MEG

Mellor, S.; O'Neill, G. C.; Bush, D.; Ramaswamy, A.; Doig, D.; Tierney, T. M.; Spedden, M. E.; Walker, M. C.; Barnes, G. R.; Vivekananda, U.

2024-10-17 neurology
10.1101/2024.10.17.24315226 medRxiv
Show abstract

Background and ObjectivesMagnetoencephalography (MEG) using optically pumped magnetometers (OP-MEG) is a relatively novel neuroimaging modality that holds great clinical promise for presurgical planning in epilepsy. However, there is limited data demonstrating that interictal discharges from deep neural sources can be visually identified and localised using OP-MEG. In this study, we sought to demonstrate the potential of OP-MEG for recording interictal epileptiform activity from patients with mesial temporal lobe epilepsy, the most common focal epilepsy, with a diverse range of aetiologies. MethodsWe recorded whole-head OP-MEG for a minimum of 30 minutes from 8 patients with temporal lobe epilepsies. Patients were seated with their head unconstrained. For comparison, we collected information from previous clinical assessment, including findings from MRI and EEG telemetry. ResultsWe observed interictal epileptiform activity in 4 of the patients, in 3 of which the localisation of the activity was concordant with previous clinical MRI and EEG. In 2 of those patients, we also observed ictal events. Of those 2 patients, one had a clear abnormality in MRI, with which the localisation of this ictal activity was concordant. The other patients MRI was negative, but the ictal localisation is consistent with the seizure semiology. DiscussionWe demonstrate that interictal and ictal epileptiform activity from mesial temporal lobe epilepsies can be observed with OP-MEG and localised to the region of the anatomical lesion, identified in MRI. This affirms the utility of OP-MEG for epilepsy surgery planning.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.