53. Evaluation of User Motor Intention Detection for Exoskeleton Control with the Muscle Cuff Regenerative Peripheral Nerve Interface (MC-RPNI) in a Rat Model

PURPOSE: Exoskeletons have been developed as promising devices for restoration of function for patients with motor extremity impairment. Despite recent significant advancements in rehabilitation robotic technology, there remains a critical need for a functional and reliable interface between user and device. Directly interfacing with the peripheral nervous system has the potential to decode motor intention, however, current peripheral nerve interfaces that utilize penetrating, intrafascicular, or extraneural cuff electrodes produce low amplitude motor action potentials and low signal-to-noise ratios (SNR), limiting their clinical application. To overcome these limitations, we have developed the Muscle Cuff Regenerative Peripheral Nerve Interface (MC-RPNI). The MC-RPNI is composed of a free, autologous muscle graft wrapped circumferentially around an intact peripheral nerve. The MC-RPNI regenerates and reinnervates through collateral sprouting from intact axons within the encompassed nerve, enabling the construct to amplify efferent motor action potentials by several magnitudes, thereby allowing for higher fidelity signaling and motor intention detection. The objective of this study was to demonstrate and correlate amplified efferent motor action potentials produced by the MC-RPNI with the gait cycle and joint angles during volitional ambulation to further evaluate the accuracy of user intention detection by this interface. METHODS: Implantation of MC-RPNI placed around the common peroneal (CP) nerve was performed in an uninjured rat model. All rats underwent treadmill ambulation training and maintenance. At three-months post surgery, animals underwent implantation of recording electrodes into the MC-RPNI. Gait patterns and joint angle data were then collected using a marker-based motion capture system during volitional ambulation on a treadmill with concurrent recording of neural signals from the MC-RPNI using EMG. RESULTS: MC-RPNI constructs remained viable over the three-month maturation period in all animal subjects. Electrophysiologic analysis conducted during volitional ambulation on a treadmill demonstrated that the MC-RPNIs amplified physiologic CP nerve signaling, generating recordable CMAPs. Furthermore, the rats’ gait cycle and joint angle data collected using a marker-based motion capture system aligned with these neural signals from the MC-RPNI constructs. CONCLUSION: The MC-RPNI is capable of amplifying neural signals from intact nerves into large, recordable EMG signals that correlate with gait cycle and joint angles during volitional ambulation on a treadmill in a rodent model. These acquired signals could be used to determine user intent and facilitate accurate exoskeleton control. Future work will focus on the utilization of amplified signals from the MC-RPNI to control a hindlimb exoskeleton. Results from these animal studies have the potential to translate into human trials with advanced exoskeleton devices for rehabilitation and treatment of extremity weakness and motor impairment.