Related Experiment Videos
Modality-specific spike identification in simultaneous magnetoencephalography/electroencephalography: a
Maeike Zijlmans1, Geertjan M Huiskamp, Frans S S Leijten
1Department of Clinical Neurophysiology, University Medical Centre, Utrecht, The Netherlands.
Summary
Simultaneous electroencephalography (EEG) and magnetoencephalography (MEG) in epilepsy patients revealed that MEG often detects more spikes than EEG. This combined approach aids in identifying and analyzing epileptic spikes, improving source modeling.
Area of Science:
- Neuroscience
- Medical Imaging
- Epilepsy Research
Background:
- Epileptiform spikes can differ in morphology and signal-to-noise ratio between electroencephalography (EEG) and magnetoencephalography (MEG).
- Lack of established criteria for MEG spikes hinders consistent identification and assessment.
- Simultaneous recordings may offer complementary information, but their combined utility for spike identification requires clarification.
Purpose of the Study:
- To investigate the consistency and reliability of visual spike assessment in simultaneous MEG/EEG data from mesial temporal lobe epilepsy (MTLE) patients.
- To compare the number and characteristics of epileptiform spikes identified in MEG versus EEG independently.
- To develop a reproducible method for selecting spikes from combined MEG/EEG data for source modeling.
Main Methods:
- Nineteen MTLE patients underwent simultaneous whole-head MEG/EEG recordings.
- EEG and MEG data were analyzed independently by three observers for spike occurrence.
- Interobserver agreement was assessed using kappa values; spikes with a mean kappa > 0.5 were considered unequivocal.
- A template match algorithm was used to compare morphology and distribution of agreed-upon spikes.
Main Results:
- Unequivocal spikes were identified in both MEG and EEG in one patient, MEG only in two, EEG only in two, and neither in 14 patients.
- MEG detected 50-80% more spikes than EEG in the index patients.
- Averaged MEG spikes showed concomitant EEG spikes, but not always vice versa.
- Simultaneous field maps revealed inconsistencies even when spikes met selection criteria.
Conclusions:
- Simultaneous MEG/EEG recordings do not automatically guarantee the best of both modalities but can improve spike identification.
- MEG, with its higher signal-to-noise ratio, can identify more spikes than EEG.
- A three-step approach to combined data analysis ensures reproducible spike selection for source modeling.
- Experienced electroencephalographers can reliably identify MEG spikes in some MTLE cases.