Genetic algorithm optimization of fuzzy outputs for classification of epilepsy risk levels from EEG signals

This paper aims to optimize the output of diagnosis of the epilepsy activity in EEG (electroencephalogram) signal by fuzzy logic techniques using genetic algorithms (GA). The fuzzy techniques are used to classify the risk levels of epilepsy based on extracted parameters like energy, variance, peaks, sharp and spike waves, duration, events and covariance obtained from the EEG of the patient. A binary GA and continuous GA are then applied on the classified risk levels to obtain the optimized risk level that characterizes the patient's epilepsy risk level. The performance index (PI) and quality value (QV) are calculated for both the method. A group of eight patients with known epilepsy findings are used for this study. High PI such as 92% (BGA) and 96% (CGA) were obtained at QV's of 80% and 90% respectively. We find that the continuous genetic algorithm provides a good tool for optimizing the epilepsy risk levels.