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

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.

Towards optimal visual presentation design for hybrid EEG—fTCD brain–computer interfaces

“…In terms of accuracy, the proposed hybrid BCI outperforms most of the methods in comparison. However, the systems introduced by Putze et al [46] and Buccino et al [49] achieved higher accuracy compared to ours as they obtained 94.70% and 94.20% average accuracy respectively. Yet, these system are slower than our system since the one introduced by Putze et al [46] requires at least 12.5 s as a task period and 20 s as a resting period while the one presented by Buccino et al [49] requires a baseline period of 6 s before starting each task.…”

Integrating EEG and MEG Signals to Improve Motor Imagery Classification in Brain–Computer Interface

“…20 These challenges critically affect the usability of MI-based BCIs 21 and have motivated, on the one hand, a deeper understanding of mechanisms associated with MI, [22][23][24][25] and on the other hand the research of new features to enhance BCI performance for both healthy subjects and patients. [26][27][28][29] In the latter case, hybrid and multimodal approaches adding respectively different type of biosignals 27,30 and neuroimaging data, such as near-infrared spectroscopy (NIRS) [31][32][33][34] and functional magnetic resonance imaging (fMRI), 35 have been proven to increase the overall performance.…”

Hemodynamic correlates of spontaneous neural activity measured by human whole-head resting state EEG + fNIRS

“…Technical progress has also made it possible to combine EEG with functional NIRS (fNIRS), another non-invasive method. This method, we refer to as EEG+fNIRS, yields similar measurements with lower space but higher time resolution, in a far more practical and cost-effective arrangement (Buccino et al, 2016;Giacometti and Diamond, 2013;Keles et al, 2014;Koch et al, 2006Koch et al, , 2008Koch et al, , 2009Roche-Labarbe et al, 2008;Safaie et al, 2013). The utility of fNIRS as an independent modality for investigating the adult brain hemodynamics (Gentili et al, 2013;Mesquita et al, 2010;White et al, 2009) as well as infant development (Lloyd-Fox et al, 2010) is already well established.…”