Integration of DWT, FFT, and Spatial domain for the identification of epileptic seizure utilizing electroencephalogram signal

Abstract Manual detection of epileptic seizure is a critical task by monitoring electroencephalogram (EEG) signals. It is also a very laborious job and requires highly trained experts. Machine learning-based approaches are widely used to solve the aforementioned problem due to their capability to detect seizures by processing long lengths of data sequences, and they also provide a high detection rate. This study provides a machine learning-based approach to detect epilepsy from the EEG signal by integrating the discrete wavelet transform (DWT), fast Fourier transform (FFT), and the Spatial domain. For the feature selection, two dimensionality reduction techniques, principal component analysis (PCA), and linear discriminant analysis (LDA) have been used. Here, 3 different classifiers, support vector machine (SVM), logistic regression (LR), and K-nearest neighbors (K-NN) have been used to diagnose epilepsy automatically. This proposed method yielded a maximum accuracy of 99% for the combination of PCA with SVM in the transform domain for the feature length of only 10.