Comparative Analysis of Electroencephalogram Signals using Ensemble methods for Epilepsy Detection

Abstract

A timely and precise diagnosis is essential for effective treatment of epilepsy, a neurological condition characterized by recurrent seizures. Because of their capacity to capture cerebral processes, electroencephalogram (EEG) data are crucial in the diagnosis of epilepsy. The proposed system gives a detailed comparison of EEG signal processing strategies for epilepsy detection utilizing ensemble techniques and investigates the usefulness of ensemble algorithms such as gradient boosting, AdaBoost, XGBoost, and bagging classifier in improving epilepsy detection. Through the use of these ensemble approaches, the system preprocesses the EEG data, extracts features, and classifies them. Accuracy, precision, recall, and F1-score are performance indicators that are used to assess each ensemble approach's efficiency. The results were obtained through extensive testing on a well-curated dataset. The finding of the proposed system clarifies the positive impacts and regulates each ensemble technique for determining the presence of epilepsy.