"Epilepsy is the second most common brain disorder after migraine; automatic detection of epileptic seizures can considerably improve the patients’ quality of life. Current Electroencephalogram (EEG)-based seizure detection systems encounter many challenges in real-life situations; EEG data are prone to numerous noise types that negatively affect the detection accuracy of epileptic seizures. To address this challenge, we propose a deep learning-based approach that learns the discriminative EEG features of epileptic seizures and to distinguish between the different types of patient recordings. More specifically, we aim to tackle this issue by using a Long Short-Term Memory network, and explore the capabilities of this model."
The data, preprocessing, model exploration, results, and conclusion are described in detail in the report ('LSTM_EpilepticSeizureRecognition.pdf').
The data is made freely available by the department of epileptology at Bonn University: https://tinyurl.com/yylxbzfj. All data points have been pre-processed, agglomerated, and correctly labeled in the all_data_epileptic_seizures.csv file.
Submitted as a final project for the 'Deep Learning' course at Nova IMS 2019.