Epilepsy Disease Detection using Artificial Neural Network and Performance Improvement using PSO and GA
Electroencephalogram (EEG) is one of the most useful clinical tools for the study of brain disorders but the
complexities involved in EEG waves make it difficult for medical practitioners to interpret actual condition of the disease. In
this study, detection of Epilepsy disease was performed using EEG signals. The classification process was applied over 13
different features extracted from online EEG dataset and performance of the neural network was evaluated on the basis of
four parameters as accuracy, precision, sensitivity, and selectivity. In order to improve Mean Square Error of the detection
process, Particle Swarm Optimization Technique and Genetic Algorithm were applied. The analysis reveals that Genetic
Algorithm can provide an efficient reduction in MSE for the proposed Epilepsy Detection System.
Keywords - Electroencephalogram (EEG) Signals, Artificial Neural Network (ANN), Particle Swarm Optimization (PSO),
Genetic Algorithm (GA), Epilepsy, Seizures, Ictal, Interictal, Mean Square Error (MSE).