Back to 2015 Annual Meeting Program
Signals from Adjacent Antagonistic RPNIs are Independent during Voluntary Walking
Andrej Nedic, MSE1; Daniel C. Ursu, MS2; Jana D. Moon, BS1; Cheryl A. Hassett, BS1; Richard B. Gillespie, PhD2; Nicholas B. Langhals, PhD1; Paul S. Cederna, MD1; Melanie G. Urbanchek, PhD1 1Department of Surgery, Section of Plastic Surgery, University of Michigan, Ann Arbor, MI; 2Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI
Introduction: Regenerative Peripheral Nerve Interfaces (RPNIs) are neurotized muscle grafts that transduce peripheral nerve action potentials to electrical signals suitable for prosthesis control. Each RPNI controls a single movement. At least two independent RPNIs are required for agonist/antagonist control; however, it is unknown whether signals from adjacent RPNIs are independent. Our purpose was to determine whether two, adjacent RPNIs, with one reinnervated by a foot dorsi-flexor nerve and the second reinnervated by a foot plantar-flexor nerve provide independent activation.
Methods: For RPNI group rats (n=2), two muscles were grafted to the left thigh and implanted with electrodes; one was neurotized with the transected tibial nerve, the other by the transected peroneal nerve. Control rats (n=2) had bipolar electrodes implanted onto soleus and extensor digitorum longus left leg muscles. Rats walked on a treadmill, were videographed and raw electromyography (EMG) signals were recorded. Rectified EMG was integrated, normalized to time, and expressed as percent of total stepping cycle activity for each stance and swing. Data are 16 stepping cycles for each rat.
Results: EMG activity for RPNI and Control rats displayed alternating patterns of activation during stance and swing (Fig 1). We found significantly greater tibial nerve activity during stance compared to swing and greater peroneal nerve activity during swing compared to stance for each nerve in RPNI and Control rats (p<0.00001) (Fig. 2). Out of 32 stepping cycles both RPNIs were active when expected 31 times—96.9% sensitivity (Fig. 2). Independence was found for RPNI activation by comparing tibial nerve reinnervated muscle activity during stance to peroneal nerve reinnervated muscle activity during stance. The strength of the significant difference indicates signaling is independent (p<0.00001). The same significant difference was found for tibial nerve reinnervated RPNI swing compared to peroneal nerve reinnervated RPNI swing (p<0.00001). Independent activity was also found for the Control group muscles.
Conclusion: Two, adjacent RPNIs, peroneal nerve reinnervated RPNI muscle and tibial nerve reinnervated RPNI muscle, were found to independently activate during quadripedal stepping cycles. These data support RPNI independence and the feasibility of using RPNIs to provide agonist/antagonist control signaling for myoelectric prostheses.
Acknowledgment: DARPA (N66001-11-C-4190).
Back to 2015 Annual Meeting Program