A Proposal to Automate Seizure Detection based on a Comparative Study of EEG Signal Analysis

Abstract

Epilepsy is a chronic neurological disorder which is characterized by recurrent and sudden seizures. People with epilepsy suffer from multiple types of seizures and Electroencephalography is an important clinical tool for diagnosing, monitoring and managing neurological disorders related to epilepsy. EEG signals are most often used to diagnose epilepsy, as seizures cause anomalies in EEG readings. In today's world where adult life expectancy is rising and humans are living longer than ever before, the healthcare system generates vast amounts of data, including EEG signals. This paper examines the prospects and challenges faced in utilizing this data in order to optimize seizure detection in order to improve the patients' quality of life. This paper also explores how Machine Learning can be applied to extract features and analyze the EEG signals and propose methods to achieve high classification accuracy.