Automatic epileptic seizure detection using LSTM networks

Abstract

Purpose

The purpose of this work is to make a computer aided detection system for epileptic seizures. Epilepsy is a neurological disorder characterized as the recurrence of two or more unprovoked seizures. The common and significant tool for aiding in the identification of epilepsy is Electroencephalography (EEG). The EEG signals contain information about the electrical activity of the brain. Conventionally, clinicians study the EEG waveforms manually to detect epileptic abnormalities, which is very time-consuming and error-prone.

Design/methodology/approach

The authors have presented a three-layer long short-term memory network for the detection of epileptic seizures.

Findings

The network classifies between seizure and non-seizure with 99.5% accuracy only in 30 epochs. This makes the proposed methodology useful for real-time seizure detection.

Originality/value

This research work is original and not plagiarized.

Citation

Shekokar, K.S. and Dour, S. (2022), "Automatic epileptic seizure detection using LSTM networks", World Journal of Engineering, Vol. 19 No. 2, pp. 224-229. https://doi.org/10.1108/WJE-06-2021-0348