Hybrid EEG-fNIRS Asynchronous Brain-Computer Interface for Multiple Motor Tasks

TLDR

Non‑invasive BCIs have shown promise for neuroprosthetics and assistive devices, and prior work has highlighted the benefits of combining EEG and fNIRS. The study aims to develop an asynchronous SMR‑based BCI that fuses EEG and fNIRS, introducing slope‑based features to promptly detect hemodynamic changes. Four motor tasks were classified using EEG and fNIRS signals processed with Common Spatial Patterns, regularized across 15 subjects and optimized via genetic algorithms, with 25 trials per class. Feature comparison demonstrates that the proposed methods reduce fNIRS delay, improving dynamic accuracy across trials.

Abstract

Non-invasive Brain-Computer Interfaces (BCI) have demonstrated great promise for neuroprosthetics and assistive devices. Here we aim to investigate methods to combine Electroencephalography (EEG) and functional Near-Infrared Spectroscopy (fNIRS) in an asynchronous Sensory Motor rhythm (SMR)-based BCI. We attempted to classify 4 different executed movements, namely, Right-Arm—Left-Arm—Right-Hand—Left-Hand tasks. Previous studies demonstrated the benefit of EEG-fNIRS combination. However, since normally fNIRS hemodynamic response shows a long delay, we investigated new features, involving slope indicators, in order to immediately detect changes in the signals. Moreover, Common Spatial Patterns (CSPs) have been applied to both EEG and fNIRS signals. 15 healthy subjects took part in the experiments and since 25 trials per class were available, CSPs have been regularized with information from the entire population of participants and optimized using genetic algorithms. The different features have been compared in terms of performance and the dynamic accuracy over trials shows that the introduced methods diminish the fNIRS delay in the detection of changes.