Title: Epileptic seizure prediction using EEG signals - a survey

Authors: Aarti Sharma

Addresses: Department of ECE, ABES Engineering College, Ghaziabad, UP, India

Abstract: Epilepsy is incurable disease of human brain that stimulates repeating seizures. A group of neurons of the human brain start firing synchronously as seizure approaches. The sudden and apparently unpredictable nature of epilepsy is one of the most disabling aspects. Recent research has explored that it is feasible to forecast the seizure in advance. Ample amount of research has been made in seizure prediction but still the representation of the current approach to the clinical application is not possible. It has been proven that seizure start with small flash of activity and takes hours to build. Therefore, automatic prediction of such activity has the potential to give precautionary warning to the patients so that harmful activities can be avoided. The present study reviews the most significant and recent methods for seizure prediction. The main difficulties in epilepsy prediction algorithms that have been identified during the preparation of this review study are feature selection and classification. The strategies presented in this study vary based on the features and classifiers used throughout the past few years. The approaches discussed will provide a thorough overview, suggestions, and directions for further research.

Keywords: seizure forecasting; algorithm; statistical validation.

DOI: 10.1504/IJBET.2025.150081

International Journal of Biomedical Engineering and Technology, 2025 Vol.49 No.3, pp.189 - 201

Accepted: 08 Jul 2025
Published online: 28 Nov 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article