Epileptic Seizure Detection: A Deep Learning Approach

• DOI: 10.48550/arxiv.1803.09848
• Publication Date: 2018-01-01
• Authors: Ramy Hussein; Hamid Palangi; Rabab Ward; Zhenzhen Wang
• Keywords: Softmax function; Epileptic seizure; Discriminative model

Epilepsy is the second most common brain disorder after migraine. Automatic detection of epileptic seizures can considerably improve patients' quality life. Current Electroencephalogram (EEG)-based seizure systems encounter many challenges in real-life situations. The EEGs are non-stationary signals and patterns vary across patients recording sessions. Moreover, EEG data prone to numerous noise types that negatively affect accuracy seizures. To address these challenges, we introduce use a deep learning-based approach automatically learns discriminative features Specifically, reveal correlation between successive samples, time-series first segmented into sequence non-overlapping epochs. Second, Long Short-Term Memory (LSTM) network used learn high-level representations normal patterns. Third, fed Softmax function for training classification. results on well-known benchmark clinical dataset demonstrate superiority proposed over existing state-of-the-art methods. Furthermore, our shown be robust noisy conditions. Compared current methods quite sensitive noise, method maintains its high performance presence artifacts (muscle activities eye-blinking) as well white noise.