Qualitative diagnostic criteria into objective quantitative signal feature classification

Predicting the epileptic seizure is challenging biomedical problem. EEG signal includes enormous information. Few relevant parameters are expected in the field of recognition and diagnostic purposes. Seizure detection and classification system has been designed and developed. The system uses computer based procedures to detect seizure and classified normal and abnormal subjects. Intelligent compact support property of GLCM is used for extracting essential features from the EEG signal. Selected features are classified using neural network model. The 70 samples would be divided into 50 training samples and 20 testing samples. The back propagation algorithm tested on these samples showed expected classification. The main objective of this study is to predict the epileptic seizure using GLCM feature extraction method and neural network model. © 2005 - 2012 JATIT & LLS.