ONLINE DETECTION OF SCALP-INVISIBLE MESIAL-TEMPORAL BRAIN INTERICTAL EPILEPTIFORM DISCHARGES FROM EEG

Brain interictal epileptiform discharges (IEDs) are transient events occurring between two or before seizure onsets. The IEDs are captured mainly by the intracranial EEG (iEEG) and only 4.7% of them are visible in our scalp EEG (sEEG) dataset. Here, we propose a method namely temporal components analysis (TCA) to detect the IEDs from ongoing sEEG and iEEG signals recorded simultaneously. In addition, spatial components analysis (SCA) previously proposed by ourselves is employed to detect the IEDs. Finally, both TCA and SCA are combined to boost the IED detection system performance. This method is referred as to TCA-SCA. The proposed TCA-SCA method detects the IEDs from the iEEG and sEEG by 81.2% and 37.4% sensitivity, respectively. The findings show that our proposed method enables the detection of scalp-invisible IEDs from ongoing sEEG recordings.