On the use of generative deep neural networks to synthesize artificial multichannel EEG signals

TL;DR

This paper reviews recent advances in using generative deep neural networks to synthesize artificial multichannel EEG signals and demonstrates feasibility with conditional variational autoencoders to produce condition-specific EEG patterns.

Contribution

It introduces a novel approach using conditional variational autoencoders to generate realistic, condition-specific multichannel EEG signals from resting-state data.

Findings

Successful generation of spectro-temporal EEG patterns for different motor imagery conditions

Feasibility of manipulating real EEG epochs to create synthetic signals

Potential applications in neural engineering and brain-computer interfaces

Abstract

Recent promises of generative deep learning lately brought interest to its potential uses in neural engineering. In this paper we firstly review recently emerging studies on generating artificial electroencephalography (EEG) signals with deep neural networks. Subsequently, we present our feasibility experiments on generating condition-specific multichannel EEG signals using conditional variational autoencoders. By manipulating real resting-state EEG epochs, we present an approach to synthetically generate time-series multichannel signals that show spectro-temporal EEG patterns which are expected to be observed during distinct motor imagery conditions.