Abstracts

Rationale: To determine if distinguishable spatial patterns are present in the stochastic behavior of phase synchronization and cross-frequency couplings on separate days in the interictal, high density scalp EEG, and if these patterns correspond to the proven epileptogenic zone. Methods: We studied one patient with refractory epilepsy who underwent intracranial EEG to establish the localization of seizure onset. Prior to invasive EEG studies the subject underwent dense array 256 channel EEG (dEEG) recordings. Three minutes of interictal dEEG data from first day, second day and on the eleventh day was randomly selected for analysis. The selected segment in each day was at least two hours temporally from an epilepic seizure and, based on visual analysis, and free of interictal epileptiform patterns. Excessively noisy channels were removed and replaced with averages of surrounding electrodes. Data were imported into MATLAB for analysis. The EEG data was filtered in the appropriate EEG band. The analysis was performed in theta (3-7 Hz), alpha (7-12 Hz), beta (12-30 Hz) and low gamma (30-50 Hz) ranges. The phase synchronization index (SI) was computed after taking Hilbert transform of the EEG data. The SI for each electrode was averaged over with the nearby six electrodes. A detrended fluctuation analysis (DFA) was used to find the stochastic behavior of the SI. The stochastic behavior of the cross-frequency coupling in theta-gamma, beta-gamma and alpha-gamma bands was also examined. Contour plots with 10 sec intervals were constructed using a montage of the layout of 256 electrode positions. Results: Contour plots displayed over the scalp show that the stochastic behavior of the SI becomes stronger with time in the proven epileptogenic area while in other areas it becomes fragmented and scattered. The low gamma band was found to be the best indicator frequency band for localization. In contrast the cross-frequency couplings decreased in the epileptogenic area for all comparisons. For this subject, we were able to identify stable patterns of stochastic behavior in the epileptogenic area within 120 seconds of analysis. Similar findings were observed in all three days of the data. Conclusions: Stochastic behavior of SI increased and the cross-frequency coupling decreased in the epileptogenic zones on all three days for this subject. These findings have the potential to assist in localizing epileptic sites in subjects with proven epilepsy.