Abstracts

Rationale: Scalp EEG is the only established non-invasive electrophysiological mapping tool that provides reliable seizure recording for pre-surgical evaluation of epilepsy. In the past decades, electrical source imaging (ESI) from interictal scalp EEG is increasingly validated and used as a valuable tool as a reflection of the irritative zone. However, ESI of ictal scalp EEG to localize the seizure onset zone (SOZ) remains challenging. Using a long-term dense-array EEG monitoring protocol along with a dynamic seizure imaging (DSI) algorithm, we were able to localize the ictal activity in concordance with surgically resected zones and ictal intracranial EEG recordings in the cohort of patients. Methods: 10 patients with focal epilepsy who had acquired T1 weighted MRI imaging with long-term 76-channel EEG recording were included in this study. Seizures were visually inspected and the electrophysiological onset were identified by experienced clinicians. A realistic geometry boundary element (BEM) model was built for each individual patient based on the preoperative MRI. The ictal EEG epochs were then analyzed using a dynamic seizure imaging technique to image the spatiotemporal seizure activity. The recordings were also down-sampled to explore the effect of different montages, i.e. 76, 32, 19 electrodes. The localization results at the onset were then compared with the surgical resection, which was extracted from post-operative MRI and the SOZ electrodes, extracted from CT images and based on physician’s reports. Results: All 10 patients underwent resective surgery and had at least 6-month follow-up. In 6 patients with known SOZ intracranial electrodes, the mean localization error between the estimation and the SOZ electrodes is approximately 1.19 cm. In the subset of patients who became seizure-free, the localization error is smaller. In 4 patients with post-operative MRI (laser ablation cases), the mean localization error between the estimation and boundary of the resection is approximately 1.79 cm, which may be due to the relatively deep foci of such cases. Conclusions: With dense-array EEG recording and DSI algorithm, we were able to localize the SOZ with reasonable accuracy in 10 epilepsy patients. The results demonstrate the capability of imaging spatiotemporal seizure sources noninvasively, which nowadays can only be achieved by invasive procedures in clinical practice. It was also shown that the localization accuracy improves when electrode numbers are increased, which suggests the benefit of using high resolution EEG recording system. Funding: NIH EB021027, NIH NS096761