AUTHOR=Gao Yunyuan , Gao Bo , Chen Qiang , Liu Jia , Zhang Yingchun 

TITLE=Deep Convolutional Neural Network-Based Epileptic Electroencephalogram (EEG) Signal Classification

JOURNAL=Frontiers in Neurology

VOLUME=Volume 11 - 2020

YEAR=2020

URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2020.00375

DOI=10.3389/fneur.2020.00375

ISSN=1664-2295

ABSTRACT=Electroencephalogram (EEG) signals contain vital information on electrical activities of brain, and are widely used to aid epilepsy analysis. As a challenging effort in epilepsy diagnosis, accurate classification of different epileptic states is of particular interest, and has been extensively investigated. A new deep learning based classification methodology, namely epileptic EEG signal classification (EESC), is proposed in the paper. This methodology first transforms the epileptic EEG signals to power spectrum density energy diagrams (PSDED), then applies deep convolutional neural networks (DCNNs) and transfer learning to automatically extract features from PSDED, and finally classifies four categories of epileptic states (interictal, preictal duration to 30 minutes, preictal duration to 10 minutes, and seizure). It outperforms the existing epilepsy classification methods in terms of accuracy and efficiency. For instance, it achieves an average classification accuracy of over 90% in the case study with CHB-MIT epileptic EEG data.