BCI-Based Human Hand Grip Control using fNIRS and sEMG

This paper presents a comparative study of Surface Electromyography (sEMG) and Functional Near Infrared Spectroscopy (fNIRS) signals for human hand grip control. sEMG provides real time data on muscle activity, while fNIRS provides insight into hemodynamics responses. A total of 20 healthy individuals participated in the experiment, where they were instructed to perform motor execution task. A multichannel continuous wave fNIRS system records the signals from the motor cortex for hand grip task and Biopac student lab system is used to acquire signal of sEMG for same task simultaneously. k Nearest Neighbors (KNN) algorithm was used as classifier for comparison of both modalities. The sEMG modality performed comparatively better than the fNIRS in terms of accuracy, with 97.28% and 85.95% for spatio-temporal features respectively. Both modalities can be used for motor cortex activity, but sEMG is more effective because it directly measures the electrical signals generated by muscle activity, offering higher temporal resolution and immediate detection of motor intentions, whereas fNIRS measures hemodynamic responses, which have a slower response time due to the inherent delay in blood flow changes following neural activity.