EEG-GAN: Generative adversarial networks for electroencephalograhic (EEG) brain signals

TL;DR

EEG-GAN introduces a stable Wasserstein GAN framework for generating realistic EEG brain signals, enabling new applications in neuroscience and brain-computer interfaces.

Contribution

The paper presents a novel EEG-GAN architecture with stabilized training and architectural insights for effective EEG time series generation.

Findings

Generated naturalistic EEG signals with high-quality metrics

Demonstrated potential for data augmentation and signal restoration

Showed feasibility of class-specific EEG signal generation

Abstract

Generative adversarial networks (GANs) are recently highly successful in generative applications involving images and start being applied to time series data. Here we describe EEG-GAN as a framework to generate electroencephalographic (EEG) brain signals. We introduce a modification to the improved training of Wasserstein GANs to stabilize training and investigate a range of architectural choices critical for time series generation (most notably up- and down-sampling). For evaluation we consider and compare different metrics such as Inception score, Frechet inception distance and sliced Wasserstein distance, together showing that our EEG-GAN framework generated naturalistic EEG examples. It thus opens up a range of new generative application scenarios in the neuroscientific and neurological context, such as data augmentation in brain-computer interfacing tasks, EEG super-sampling, or restoration of corrupted data segments. The possibility to generate signals of a certain class and/or with specific properties may also open a new avenue for research into the underlying structure of brain signals.