EGAN: an ensemble adversarial network for topology-preserving EEG data generation for predicting cognitive load levels

Abstract Learning representations and extracting meaningful patterns from electroencephalogram (EEG) recordings is critical for analyzing cognitive events (e.g., predicting cognitive load). The primary challenges include individual variability, technical noise from unreliable sensor-skin contacts, and rapid temporal changes in the EEG recordings. Given the multi-factorial nature of the problems, deep learning models are natural choices for learning representations from the data. However, the extensive time required for data collection limits the number of subjects (samples) available, which is essential for building robust deep learning models. We introduce an ensemble generative adversarial network (EGAN) to generate high-fidelity EEG data. The EGAN generates multichannel EEG recordings and their spatial-spectral representation. Key design constraints were preserving topological structure and maintaining proper bias-variance trade-offs for building robust models. To ensure the quality of the synthetic data, we visually inspected the data generated by EGAN. We conducted spectral analyses to confirm that the quality and spectral similarity were comparable to EEG recordings. To illustrate the efficacy of data generated by EGAN, we developed a convolutional neural network (CNN) model to predict four levels of cognitive load. We used spatial-spectral representations (Topomap) from three frequency bands (i.e., $$\theta $$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mi>θ</mml:mi> </mml:math> , $$\alpha $$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mi>α</mml:mi> </mml:math> , and $$\beta $$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mi>β</mml:mi> </mml:math> ). We trained our model on real data and a mixture of original and EGAN-generated data with varying proportions (e.g., 50%, 70%, 100%) of real data for training. We compared the performance of models trained solely on original data to those trained on mixed data. Our results indicate that CNN models trained on EGAN-supplemented data significantly outperformed those trained using only real data across all three frequency bands and all training set proportions. For example, the model trained on the mixed data with 50% real data achieved classification F1 and accuracy scores of approximately 71%, 69%, and 72% for $$\theta $$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mi>θ</mml:mi> </mml:math> , $$\alpha $$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mi>α</mml:mi> </mml:math> , and $$\beta $$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mi>β</mml:mi> </mml:math> bands, respectively. The same model trained on only real data achieved approximately 52%, 57%, and 59% accuracy and F1 scores for the respective bands. These results demonstrate the utility of EGAN in generating high-fidelity EEG data, significantly advancing cognitive load analysis and modeling other cognitive events.