Epilepsy is one of the most common disorders of the brain. Currently, studies of epileptic seizures often involve tedious and time-consuming visual inspection of multi-channel long EEG data by medical experts. To better monitor seizures and make medications more effective, we propose a recurrence time based approach to characterize brain electrical activity. Unlike many other nonlinear methods, the proposed approach does not require that the EEG data be chaotic and/or stationary. It only contains a few parameters that are largely signal-independent, and hence, is very easy to use. The method detects epileptic seizures with accuracy close to 100% (when subclinical seizures are not counted) and false alarm rate per hour close to 0. Most critically, the method is very fast: with an ordinary PC (CPU speed less than 2 GHz), computation of the recurrence time from one channel EEG data of duration one hour with sampling frequency of 200 Hz takes about 1 minute CPU time. Therefore, with an ordinary PC, the method is able to process all 28 channels of 1-hour EEG data in about half an hour, and thus faster than the data being continuously collected. The method can also effectively monitor propagation of seizures in the brain. Therefore, it has the potential to be an excellent candidate for real-time monitoring of epileptic seizures in a clinical setting.

This content is only available via PDF.
You do not currently have access to this content.