Minimum Relative Entropy (MRE) Method for Fuzzy based Classification of Epilepsy Risk Levels from EEG Signals

The purpose of this research is to investigate the feasibility of minimum relative entropy (MRE) information distance in optimization of fuzzy outputs for the classification of epilepsy risk levels from EEG (electroencephalogram) signals. The fuzzy pre classifier is used to classify the risk levels of epilepsy based on extracted parameters like energy, variance, peaks, sharp and spike waves, duration, events and covariance from the EEG signals of the patient. minimum relative entropy (MRE) (post classifier with KL distance) is applied on the classified data to identify the optimized risk level (singleton) that characterizes the patient´s epilepsy risk level. 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 ten patients with known epilepsy findings are used for this study. High PI such as 96.56% was obtained at QV´s of 23.02 in the MRE optimization when compared to the value of 40% and 6.25 through fuzzy classifier respectively. We noted that the MRE (Kullback Leibler-KL) information distance is a better tool for optimization of epilepsy risk levels.