Analysis of SVD neural networks for classification of epilepsy risk level from EEG signals

Abstract

The Electroencephalogram (EEG) is a complex signal that indicates the electrical activity of brain. EEG is a signal that represents that effect of the superimposition of diverse processes in the brain. Epilepsy is a common brain disorder. Out of hundred one person is suffering from this problem. Here we study a novel scheme for detecting epileptic seizure and classifying the risk level from EEG data recorded from Epileptic patients. EEG is obtained by International 10-20 electrodes system. Singular Value Decomposition (SVD) is used for feature extraction. The efficacy of the above methods is compared based on the bench mark parameters such as Performance Index (PI), and Quality Value (QV). A group of twenty patients with known epilepsy findings are analyzed. It was identified that Elman neural network is a good post classifier in the optimization of epilepsy risk levels. © 2013 Springer.