Hybrid brain-computer interface based on EEG and NIRS modalities

Non-invasive brain-computer interfaces (BCIs) allow users to control external devices by their intentions. Currently, most BCI systems are synchronous, which means, they rely on cues or tasks to which a subject has to react. It would be more useful for users if they could control a device at their own will (i.e., asynchronous BCIs). However, previous asynchronous BCI systems that rely on non-invasive electroencephalogram (EEG) measurements, are not accurate and stable enough for real world applications. Previously, hybrid BCI systems, relying on simultaneous EEG and near-infrared spectroscopy (NIRS) measurements, have been shown to increase the classification performance of synchronous motor imagery (MI) tasks. In this study, we present a first report on an asynchronous multi-modal hybrid BCI, based on simultaneous EEG and near-infrared spectroscopy (NIRS) measurements and propose novel subject-dependent classification strategies for combining both measurements.