Activation of paralyzed muscles with electrical stimulation enables people with paralysis to perform lost functions and improve their overall health through exercise. Lower extremity stimulation systems can evoke multiple movement patterns, including standing, stepping, cycling, and rowing. Though some functional and physiological improvements have been seen with existing systems, they are still limited by rapidly induced fatigue of the activated musculature.
Implanted neuroprostheses have recently incorporated novel multi-contact nerve cuff electrodes (NCEs) which directly surround a proximal peripheral nerve with multiple independently controlled electrode contacts. These high density interfaces open new possibilities for selectively interacting with the neural anatomy. Selective stimulation of independent yet synergistic motor unit pools may be instrumental in alleviating the pervasive issue of rapid fatigue.
We herein investigate several selective stimulation strategies through multi-contact NCEs and assess their ability to prolong activated muscle output. Three main strategies include decreased fiber overlap, decreased duty cycle, and feedback control of stimulus intensity. Pre-clinical testing narrowed the activation period and duty cycle parameter spaces to those that best maintained joint moment output for different fiber type classifications. These early investigations also verified the feasibility and advantages of controlling selective neural stimulation intensity with functional feedback.
Pre-clinical findings were translated to human neuroprosthesis recipients with lower extremity paralysis. Clinical studies began with isometric analyses of two feedback-controlled duty cycle reducing paradigms: Carousel and Sum of Phase Shifted Sinusoids (SOPS) stimulation. Both paradigms, implemented with insights on activation period, duty cycle, and feedback from pre-clinical results, significantly prolonged functionally-relevant isometric knee joint moments in one participant with spinal cord injury (SCI).
We next investigated selective stimulation strategies during dynamic cycling exercise. Six participants with paralysis cycled with low fiber overlap, low duty cycle, and/or cadence controlled stimulation. At least one selective strategy significantly improved endurance and sustained exercise intensity over conventional techniques in five participants. We further demonstrated significant physiological improvements from muscle oxygenation and heart rate analyses. We ultimately find that selective stimulation strategies through NCEs are successful in prolonging activated muscle output and improving exercise performance.