TUC participated in the IEEE EMBC 2025 conference, presenting the paper “Enhancing EEG Classification for Motor Imagery Control of a VR Game based on Deep Learning Techniques on Small Datasets”. In this work, a Wasserstein Generative Adversarial Network (WGAN) was implemented to augment small datasets with artificial EEG features. A shallow Convolutional Neural Network (CNN) was also built to effectively eliminate noise and extract robust representations of the feature space. The system was evaluated in a VR maze game, and the results are promising: higher accuracy with minimal overfitting for naive BCI users compared to linear- and Riemann-based machine learning algorithms.
TUC at the 2025 IEEE Engineering in Medicine and Biology Society (EMBC2025)
