Epileptic Seizure Prediction Based on Localization

Abstract

Electroencephalography (EEG) gives complex information about the brain dynamics. Visual Inspection of EEG signal is difficult and requires skilled interpreters in identifying epileptic seizures. One third of the people having epilepsy are drug resistant and the latest anti-epileptic drugs cannot stop the seizures completely. Unexpected occurrence of seizure disturbs the quality of life and causes physical damage. The Prediction based on localization of Epileptic Zone using EMD and neural network has resulted excellent prediction accuracy. A seizure prediction time of 30.4 minutes was achieved. It can be used as a clinical application for the diagnosis and effective epilepsy management.