Abstracts

Rationale: EEG source imaging is a useful tool to localize epileptogenic foci because of its high temporal resolution at the millisecond scale and its noninvasive nature. A novel ictal source analysis approach was previously developed in our lab for spatial-temporal source localization and connectivity analysis. In this study, we evaluate the ability of the approach to identify epileptogenic foci for high-density EEG recordings in partial epilepsy patients. Methods: High-density EEG seizure data with 76 channels was recorded in eight medically intractable partial epilepsy patients. All the patients had surgical resections, and were either seizure free or had significant seizure reduction one year after the operation. The EEG sources were modeled as the equivalent current dipoles. Patient-specific boundary element head models were created from their structural MRI. Subspace source imaging method (FINE) was used to localize the ictal sources in the 3-Dimensional brain. Source space connectivity analysis approach was then applied to identify the primary ictal sources. The results were evaluated by comparing the source locations with surgical resections. Results: 18 seizures in eight patients were analyzed in this study. Source locations of 11 seizures were within the surgical resection lesions. Source locations of 4 seizures were within 10 mm distance to the lesion boundaries. Sources in the remaining 3 seizures were located in brain areas more than 10 mm away from surgical resection lesions. Source analysis for these eighteen seizures had a mean localization error 4.02 mm and standard deviation 6.93 mm. Conclusions: High-density EEG source imaging and connectivity analysis approach were applied to study the noninvasive ictal source localization. The source imaging results were well co-localized with the post-operative MRI lesions. Majority of the ictal sources identified were either overlapping with, or close to patient surgical resections. These results suggest the potential application of high-density EEG source imaging in localizing epileptogenic foci for pre-surgical planning.