Compact Convolutional Neural Networks for Multi-Class, Personalised, Closed-Loop EEG-BCI

TL;DR

This paper introduces a compact CNN-based BCI system enabling users with motor disabilities to switch between four control modes in real-time using EEG signals, demonstrating promising online classification performance in a real-world gaming scenario.

Contribution

The study presents a novel, efficient CNN architecture (SmallNet) for multi-class, self-paced EEG classification, optimized for real-time, user-friendly BCI control in domestic environments.

Findings

Online trained models outperform offline models in real-time prediction.

The system achieves an online accuracy of 47.6% in a real-time game setting.

The approach doubles the number of control states compared to previous physiological signal decoders.

Abstract

For many people suffering from motor disabilities, assistive devices controlled with only brain activity are the only way to interact with their environment. Natural tasks often require different kinds of interactions, involving different controllers the user should be able to select in a self-paced way. We developed a Brain-Computer Interface (BCI) allowing users to switch between four control modes in a self-paced way in real-time. Since the system is devised to be used in domestic environments in a user-friendly way, we selected non-invasive electroencephalographic (EEG) signals and convolutional neural networks (CNNs), known for their ability to find the optimal features in classification tasks. We tested our system using the Cybathlon BCI computer game, which embodies all the challenges inherent to real-time control. Our preliminary results show that an efficient architecture (SmallNet), with only one convolutional layer, can classify 4 mental activities chosen by the user. The BCI system is run and validated online. It is kept up-to-date through the use of newly collected signals along playing, reaching an online accuracy of 47.6% where most approaches only report results obtained offline. We found that models trained with data collected online better predicted the behaviour of the system in real-time. This suggests that similar (CNN based) offline classifying methods found in the literature might experience a drop in performance when applied online. Compared to our previous decoder of physiological signals relying on blinks, we increased by a factor 2 the amount of states among which the user can transit, bringing the opportunity for finer control of specific subtasks composing natural grasping in a self-paced way. Our results are comparable to those shown at the Cybathlon's BCI Race but further improvements on accuracy are required.