Chaos analysis of electro encephalography and control of seizure attack of epilepsy patients.

In order to evaluate the EEG of patients with epilepsy, chaos analysis was performed for the subdural EEG time series data. The chaos attractor was reconstructed in the phase space and the correlation dimension. KS entropy calculated from the Lyapunov exponents was evaluated. Before the seizure attack, the KS entropy showed a lower value when compared with the time series data recorded during healthy condition. The results of our study suggest that it is possible to predict the seizure attack by the chaos analysis of the EEG signal. Further, we aim at developing an automatic control system for predicting a seizure attack by the use of local cooling of the focus with Peltier elements.