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

• DOI: 10.1371/journal.pone.0146610
• Publication Date: 2016-01-05T19:15:58Z
• Authors: Alessio Paolo Buc...; Hasan Onur Keles; Ahmet Omurtag
• Keywords: Sensorimotor rhythm; Functional near-infrared spectroscopy; Motor Imagery

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