A SURVEY ON THE CLASSIFICATION TECHNIQUES FOR SEIZURE PREDICTION FROM EEG

Electronics and Communication Engineering

NATIONAL CONFERENCE ON TECHNOLOGICAL ADVANCEMENTS IN ENGINEERING

Epilepsy is basically a neurological disorder which is characterized by epileptic seizures. These seizures can either be partial or generalized. Epileptic seizures are sometimes accompanied by heavy and vigorous shivering or shaking which can lead to physical injuries and unconsciousness. It can even get fatal if not properly attended. Timely Seizure prediction is important as far as the safety and health of an epileptic person is concerned. It helps in taking precautions like brain stimulation or appropriate medication before a seizure strikes a person. Here comes the relevance of an efficient and accurate seizure prediction algorithm. There are several classification techniques which can be used to predict the onset of seizure. This paper primarily focuses on a review of such available methods and their important features. Classification techniques such as the Support Vector Machines, Bayesian Classifier, Artificial Neural Networks, Adaboost, Adaptive Neuro Fuzzy Inference System etc. are being discussed here. It also discusses the pros and cons of each of them. The primary aim is to conquer a classification method which can effectively be used in predicting an ictal event. Different methods use different techniques of classification which can uniquely identify the onset of a seizure thereby reducing false positives.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.