Sub-band mean phase coherence for automated epileptic seizure detection

Abstract

This paper presents the results of our study on sub-band phase synchronization of neural activity for automated detection of epileptic seizures. Mean phase coherence (MPC) as a measure of phase synchronization is extracted from the two adjacent intracranial electroencephalogram (iEEG) recordings, which are bandpass-filtered within the desired frequency bands. A threshold-based classifier is applied to generate the alarms. Results can be useful in two ways: for automated seizure detection, and for gaining better understanding of the synchronization behavior of epileptic seizures. The proposed method was applied on 5 invasive recordings selected from the EPILEPSIAE database. The results are compared with the MPC measures extracted from artifact removed raw iEEG signals, and show significant improvement for automated epileptic seizure detection. On average the subband MPC results in a sensitivity of 79% (43 of 54 seizures in 780h recordings) and a false detection rate of 0.05 per hour.