Abstracts

RATIONALE: Non-invasive imaging techniques play a major role in the diagnosis and surgical planning for patients with intractable partial epilepsy. High temporal resolution cortical mapping of the generators from the combined interictal Magnetoencephalogram (MEG) and EEG can help unveil the spatio-temporal dynamics of epileptogenic networks.
METHODS: 10 patients were concurrently recorded with a whole-head MEG system (with 151 sensors) and 67 EEG channels. A set of interictal spikes was recorded during 8 5-minute sessions while patients were asked to rest. High-resolution MR scans of the patients heads were subsequently acquired and accurate co-registration with the MEG/EEG was achieved by fitting their digitized head shapes to some scalp segmentation obtained from the MR. Patients were all suffering from intractable temporal lobe epilepsy, their anatomical MR revealing right or left hippocampal sclerosis.
Characterization of MEG/EEG generators in terms of spatial distribution, temporal dynamics and clustering across the individual spikes was achieved with new distributed cortical source imaging techniques using MEG, EEG and an optimized combination of MEG and EEG recordings (Baillet et al., 1999; Mosher et al, 1999; Gavit et al., 2000).
RESULTS: For each patient, analysis of the cortical maps for each spike revealed significant activation distributed in a network that included the afflicted hippocampus (in 47% to 98% of the spikes across patients), and also distant circumvolutions of the corresponding temporal lobe, the insula and the orbito-frontal cortex. These findings were very much in agreement with the clinical report and depth-recordings, when available.
In some patients, a stable dynamic network involving the neocortex and deeper structures could be revealed along the spike-wave complexes. We will discuss how the combination of concurrent MEG and EEG recordings improved the characterization of their generators.
CONCLUSIONS: Combined cortical mapping to the individual MR and non-linear source estimation techniques improve the characterization of the areas involved in interictal activity, both in terms of spatial resolution and in terms of the temporal dynamics of the MEG/EEG generators of the spike-wave complexes.
Baillet, S., Garnero, L., Marin, G, Hugonin J.P., IEEE Trans Biomed Eng 1999 (46). : 522-34.
Mosher, J. C., Baillet, S., Leahy, R. M., J Clin Neurophys. 1999 (16). : 225-38.
Gavit L., Baillet S., Mangin J.F., Pescatore J.& Garnero L.,Proc. Int Conf. BIOMAG2000, Aug. 2000, 717-20