SOFT THRESHOLDING TECHNIQUES WITH PCA AS POST CLASSIFIER FOR EPILEPSY RISK LEVEL CLASSIFICATION

Abstract

Due to the frequent electrochemical impulses in the neurons, recurrent and rapid disturbances of mental functions occur which is known as epilepsy and it has a drastic effect on the lives of human beings. The excessive discharge is termed as epileptic spikes which help in diagnosis and analysis of epilepsy. Electroencephalography (EEG) serves as a vital clinical tool for the analysis of epilepsy. Because of its self adaptive nature, Artificial Neural Network (ANN) has been widely used to classify the epilepsy risk levels from EEG Signals. This paper includes the concept of Soft Thresholding (ST) techniques followed by the implementation of Principal Component Analysis (PCA) for the Classification of Epilepsy Risk Levels from EEG Signals. The analysis is done in terms of benchmark parameters such as Performance Index (PI), Quality Values (QV), Sensitivity, Specificity, Time Delay and Accuracy.