Automatic detection of epilepsy seizure in EEG using entropy features and CNN classifier: A deep learning approach

Abstract

Discrete wavelet Transform (DWT) is a powerful tool that is beaing widely used in signal analysis like electroencephalography(EEG) for detetcion of epilepsy seizures. EEG  is a non invasive and continuous time signal that is recorded by placing the electrodes over the scalp. EEG signals are contaminted with unwanted noise, artifacts, etc hence, the signals need to be filtered before analysis to avoide complication. This research work proposes a novel approach for epilepsy seizure detection using a combination of Discrete Wavelet Transform (DWT) and Convolutional Neural Network (CNN) model. Electroencephalogram (EEG) signals are preprocessed using DWT to obtain multi-resolution coefficients, which capture both time and frequency domain information. Extracted features from DWT are elelction using feature selection by random forest algorithm and further coefficients are fed into a CNN. Experimental results demonstrate the effectiveness of the proposed method, showcasing high accuracy in seizure detection. The DWT-CNN model offers a promising solution for real-time and accurate epilepsy seizure detection, holding significant potential for clinical applications. Analysis of the presented algorithm is performed on the benchmark Bonn EEG dataset. The  proposed model achieves an accuracy of 99.9%, a sensitivity of 100%, a precision of 99.81%, and a specificity of 99.8%.