Performance Analysis of Support Vector Machine (SVM) for Optimization of Fuzzy Based Epilepsy Risk Level Classifications Using Different Types of Kernel Functions from EEG Signal Parameters.

In this paper, we investigate the optimization of fuzzy outputs in the classification of epilepsy risk levels from EEG (Electroencephalogram) signals. The fuzzy techniques are applied as a first level classifier to classify the risk levels of epilepsy based on extracted parameters which include parameters like energy, variance, peaks, sharp spike waves, duration, events and covariance from the EEG signals of the patient. Support Vector Machine (SVM) may be identified as a post classifier on the classified data to obtain the optimized risk level that characterizes the patient's epilepsy risk level. Epileptic seizures result from a sudden electrical disturbance to the brain. Approximately one in every 100 persons will experience a seizure at some time in their life. Some times seizures may go unnoticed, depending on their presentation which may be confused with other events, such as a stroke, which can also cause falls or migraines. Unfortunately, the occurrence of an epileptic seizure seems unpredictable and its process is very little understood The Performance Index (PI) and Quality Value (QV) are calculated for the above methods. A group of twenty patients with known epilepsy findings are used in this study. High PI such as 98.5% was obtained at QV's of 22.94, for SVM optimization when compared to the value of 40% and 6.25 through fuzzy techniques respectively. We find that the SVM Method out performs Fuzzy Techniques in optimizing the epilepsy risk levels. In India number of persons are suffering from Epilepsy are increasing every year. The complexity involved in the diagnosis and therapy is to be cost effective in nature. This paper is intended to synthesis a cost effective SVM mechanism to classify the epilepsy risk level of patients.