Automated Decision Support System for Focal Epilepsy Detection using Electroencephalogram

Epilepsy is a neurological disorder related to brain, which can be diagnosed using Electroencephalogram (EEG). The EEG consists of various rhythms that provide information on distinct brain activities. This paper proposes a methodology for the automated classification of focal and non-focal EEG signals. Wavelet packet decomposition (WPD) is used to obtain different frequency bands and further various activities such as delta, theta, alpha, beta and gamma. The different features such as energy, mean, highest magnitude, standard deviation, sample entropy, approximate entropy and Shannon entropy values are computed for each of the brain activities. The proposed methodology is demonstrated using the publicly available Bern Barcelona database. Machine learning algorithms such as Decision tree (DT), K-nearest neighbors (KNN) and Support Vector Machine (SVM) are used for the performance evaluation. SVM provided higher performance of 96.96% in comparison with other classifiers. Apart from methodological contribution, the developed system can be used as an assisted tool for the physician in his clinical practice. The results obtained demonstrate that the proposed method outperforms the available approaches.