Interictal spike localization for epilepsy surgery using magnetoencephalography beamforming.
Rui Li1, Chris Plummer2, Simon J Vogrin2
1Department of Biomedical Engineering, The University of Melbourne, Parkville, VIC, Australia.
Summary
Automated magnetoencephalography (MEG) beamforming effectively predicts seizure freedom after epilepsy surgery. This non-invasive method aids in localizing focal epilepsy and improving surgical outcomes.
Area of Science:
- Neuroscience
- Medical Imaging
- Epileptology
Background:
- Magnetoencephalography (MEG) is crucial for localizing focal epilepsy.
- Manual analysis of MEG source activity is time-consuming for neurophysiologists.
- Automated methods are needed to improve efficiency and accuracy.
Purpose of the Study:
- To develop and evaluate an automated MEG kurtosis beamforming framework for focal epilepsy localization.
- To assess the framework's effectiveness against surgical resections and patient outcomes.
- To determine if automated MEG can predict postoperative seizure freedom.
Main Methods:
- Retrospective analysis of MEG recordings from 13 epilepsy surgery patients.
- Application of kurtosis beamforming to identify regions of Maximum Interictal Spike Frequency (MISF).
- Validation of MISF regions against sLORETA modeling and surgical resection data.
Main Results:
- Successful spike localization in 12 out of 13 patients.
- Concordance between beamforming MISF regions and surgical resections in 8 patients with Engel I outcomes.
- Localization of spike onset and propagation in relation to sLORETA models.
Conclusions:
- Automated MEG beamforming can predict postoperative seizure freedom at the lobar level.
- The method tends to localize propagated MEG spikes.
- MEG beamforming shows potential for non-invasive prediction of surgical outcomes in drug-refractory focal epilepsy.
