Methods for Seizure Detection and Prediction: An Overview

Abstract

Epilepsy is one of the most common neurological diseases and the most common neurological chronic disease in childhood. Electroencephalography (EEG) still remains one of the most useful and effective tools in understanding and treatment of epilepsy. To this end, many computational methods have been developed for both the detection and prediction of epileptic seizures. Techniques derived from linear/nonlinear analysis, chaos, information theory, morphological analysis, model-based analysis, all present different advantages, disadvantages, and limitations. Recently, there is the notion of selecting and combining the most robust features from different methods for revealing various signals’ characteristics and making more reliable assumptions. Finally, intelligent classifiers are employed in order to distinguish epileptic state out of normal states. This chapter reviews the most widely adopted algorithms for the detection and prediction of epileptic seizures, emphasizing on information theory based and entropy indices. Each method’s accuracy has been evaluated through performance measures, assessing the ability of automatic seizure detection/prediction.