Detection of Pre-Epileptic Seizure by Using Wavelet Packet Decomposition and Artifical Neural Networks

Abstract

Epilepy is one of the best known disease in the world and it affects the life of patiens in a negative way. There are a lot of studies which include EEG signals are in progress to remove these problems. Accurate and reliable EEG signals give important information about the situation of the brain and its electrical activity so they are vital for epileptic seizure detection. In this study healthy and interictal EEG signals are used to detect pre-epileptic seizure. The proposed study includes three stages. Firstly, in the preprocess stage EEG signals are normalized and then wavelet packet decomposition method is used to each signal to obtain key features. At the final stage artifical neural networks are applied to classify these key features. In our study we had 99% classification rate for 100 healthy and 100 interictal EEG signals.