A Comparison of Genetic Algorithm & Neural Network (MLP) In Patient Specific Classification of Epilepsy Risk Levels from EEG
Signals –IEEE- – MATLAB

Abstract—This paper is aimed to compare the performance of a
Genetic Algorithm (GA) and Multi- Layer Perceptron (MLP)
Neural network in the classification of epilepsy risk level from
Electroencephalogram (EEG) signal parameters. The epilepsy risk
level is classified based on the extracted parameters like energy,
variance, peaks, sharp and spike waves, duration, events and covariance from the EEG of the patient. A Binary Coded GA (BCGA) and MLP Neural network are applied on the code converter’s classified risk levels to optimize risk levels that characterize the patient. The Performance Index (PI) and Quality Value (QV) are calculated for these methods. A group of ten
patients with known epilepsy findings are used in this study. High
PI such as 93.33% and 95.83% for BGA and MLP are obtained at
QV of 20.14 and 21.59.