A wireless wearable sEMG and NIRS acquisition system for an enhanced human-computer interface

Surface electromyography (sEMG) is extensively explored in human-computer interface (HCI); complementary to the electrophysiological activity of the muscles, the hemodynamic information that measured from near infrared spectroscopy (NIRS) is less investigated. Properly combining the sEMG and NIRS would provide a novel approach for HCI applications. This paper presents a multi-channel wireless wearable sEMG and NIRS acquisition system aiming for enhanced human-computer interaction, by providing more information about the muscle activity for subject's motor intention decoding. Extensive tests were carried out to evaluate the system performance. It showed that this novel system proved to be able to capture sEMG signals similar to those of the commercialized sEMG acquisition devices, and had a comparable NIRS sensor performance. Furthermore, simultaneously recording of sEMG and NIRS signals, the system had shown the ability to provide more information about the muscle activities for a better HCI performance. The classification accuracy of 13 hand gesture motions was significantly (P<;0.001) improved by using combined sEMG and NIRS features comparing to sEMG or NIRS features individually, suggesting that the proposed sEMG and NIRS system could be potentially available for an enhanced HCI.