Abstracts

Rationale: Epilepsy is characterized by recurrent, unprovoked seizures and affects about 50 million people worldwide. Nearly 80% of people with epilepsy live in low- and middle-income countries and 75% of epilepsy patients living in low-income countries do not get optimal treatment. Continuous EEG monitoring can detect seizures in real time. At present, only a trained professional can interpret the EEG leading to intermittent seizure detection. The purpose of this research was to devise a more continuous and efficient method (SeizureSeeker) for analyzing EEG data using machine learning algorithms that enable complex data processing to distinguish between normal EEG and electrographic seizures.

Methods: We used an open access EEG dataset containing pre-identified records of 500 patients and compared 3 different classification algorithms: Logistic Regression, Support Vector Machines (SVM), and Long Short-Term Memory (LSTM). All models were fitted using existing software from Python libraries and the Orange data mining application.

Results: Logistic regression had poor accuracy, but SVM achieved impressive results with an overall accuracy of 94%. LSTM is a more complex algorithm based on recurrent neural networks and generated near perfect classification results with an accuracy of 99%.

Conclusions: The memory property of the LSTM model makes it an ideal choice for time series EEG data. The LSTM is an accurate and effective machine learning model that can be built upon to generate automated seizure detection software. Such advancements can facilitate more timely diagnoses of seizures and can be used where access to specialized medical expertise is limited.

Funding: Please list any funding that was received in support of this abstract.: n/a.