Subject-dependent classification for robust idle state detection using multi-modal neuroimaging and data-fusion techniques in BCI

Abstract Brain-computer interfaces (BCIs) allow users to control external devices by their intentions. Currently, most BCI systems are synchronous. They rely on cues or tasks to which a subject has to react. In order to design an asynchronous BCI one needs to be able to robustly detect an idle class. In this study, we examine whether multi-modal neuroimaging, based on simultaneous EEG and near-infrared spectroscopy (NIRS) measurements, can assist in the robust detection of the idle class within a sensory motor rhythm-based BCI paradigm. We propose two types of subject-dependent classification strategies to combine the information of both modalities. Our results demonstrate that not only idle-state decoding can be significantly improved by exploiting the complementary information of multi-modal recordings, but also it is possible to minimize the delay of the system, caused by the slow inherent hemodynamic response of the NIRS signal.