Creating synthetic EEG through training Generative Adversarial Networks (GAN) on patient data.
Examples are shwon below, first one being the GAN model trained on normal EEG data and the one below for a model trained on data when a seizure is taking place.
The plot for normal data
The plot for abnormal data
