Neural Networks Design for Classification of Epilepsy EEG Signals

J. D. Dhande, Dr. S. M. Gulhane

Abstract: s Epilepsy is the brain disorder. It is characterized by a sudden and recurrent malfunction of the brain which is termed as seizure. The Electroencephalogram (EEG) signal is widely used clinically to investigate brain disorder. This paper investigates the application of Artificial Neural Network models for classification of epileptic EEG signals. In this work, the design approach focused to increase the number of hidden neurons and find the optimal value of hidden neurons for neural network model. The proposed work performed in two stages Initially a feature extraction scheme using statistical measures has been applied and different ANN models with back propagation learning based algorithm classifiers have designed to performed the classification. The performance of the proposed method has been studied using a publically available benchmark database [11] in terms of training performance and classification accuracy. The proposed method provides good classification accuracy and generalization characteristic.

Keywords: Epilepsy, BPNN, Electroencephalogram EEG, Artificial Neural Network ANN

How to Cite?: J. D. Dhande, Dr. S. M. Gulhane, "Neural Networks Design for Classification of Epilepsy EEG Signals", Volume 4 Issue 5, May 2015, International Journal of Science and Research (IJSR), Pages: 1249-1252, https://www.ijsr.net/getabstract.php?paperid=SUB154363, DOI: https://dx.doi.org/10.21275/SUB154363