A Comparative Study on Epileptic Seizure Detection Methods

摘要

Epilepsy is a chronic brain disease that affects around 50 million people worldwide. This disease is characterized by recurrent seizures, which are brief episodes of involuntary movement that may involve a part of the body (partial) or the entire body (generalized) and are sometimes accompanied by loss of consciousness. Seizure episodes are a result of excessive electrical discharges in a group of brain cells. Different parts of the brain can be the site of such discharges. Seizures can vary from the briefest lapses of attention or muscle jerks to severe and prolonged convulsions. Seizures can also vary in frequency, from less than 1 per year to several per day. Seizure prediction systems can be life changing for patients with epileptic seizures. By accurately identifying the periods in which seizure occurrence has a higher chance of happening we can help epileptic patients live a more normal life. In this paper we aimed to find supervised machine learning (ML) algorithms to predict the risk of seizure happening. We trained multiple classifiers including some pre-trained models using both time and frequency domain predictors from the intracranial electroencephalogram (iEEG) signals. The results are compared the performance measures in this study on a case by case basis. Our algorithm can be easily implemented in a wearable seizure warning device in conjunction with an implantable iEEG sensor. A hand-held personal advisory device can alert the patient of a possible epileptic seizure.