A prosthesis is conventionally controlled and monitored by recording and converting the electrical signals from the muscles (electromyography) into machine-understandable signals. However, the activity of the muscle can be registered not only electrically, but also magnetically, as every electrical activity within the muscle also generates a magnetic field. This magnetomyography originates from the same ionic currents and presents comparable temporal and spectral profiles, but poses two potential advantages over electromyography: a) the intrinsic contactless nature of the measurement and b) a theoretical superiority when it comes towards the decomposition of muscle signals. MMG can be recorded contactless as the magnetic permeability of human tissue is comparable to that of empty space and thereby magnetic field is less distorted. This suggests a possible superiority of MMG for the study of muscle signal and theoretically leads (in silico) to a twofold discrimination of motor neurons of the spinal cord, when compared to electromyography. Despite these advantages, current challenges are the need for magnetic shielding and sensor size and sensitivity - which can be partly surpassed using nitrogen-vacancy diamond magnetometers. Given the circumstance that little is known for magnetomyography and prosthesis control, we performed several experiments using commercially available optically pumped magnetometer to provide a reference about current technological possibilities (e.g., sensor sensitivity); furthermore, we present a new fully-integrated nitrogen vacancy-based sensor system, which may augment and/or replace traditional biosignal interfaces to peripheral devices such exoprostheses or exoskeletons.
|