PURPOSE: The Osseointegrated Neural Interface (ONI) represents a novel approach to peripheral nerve interfacing, with the potential to enable closed loop control of prosthetic limbs. By implanting electrodes within the medullary canal of long bones, the ONI may provide a stable platform for recording neural signals and delivering stimulation. This study seeks to create a clinically translatable large animal model for chronic sensory input in a freely ambulating animal.
METHODS: One female adult sheep underwent a forelimb amputation followed by surgical implantation of the ONI system. Neural electrodes were implanted within the radial nerve. Channels 2 and 4 were assigned for neural recordings, Channel 1 served as the common reference, and Channel 5 was used to record the EMG signal. Following implantation, afferent electrical stimulation trials were performed to verify whether the ONI could evoke and record afferent compound nerve action potentials (CNAPs). Graded stimulation currents were applied, and data was collected with different frequencies and amplitudes. The electrophysiological signals recorded underwent processing to subtract the reference and EMG channels from the neural channels to isolate the neural activity.
RESULTS: Afferent CNAPs were recorded from Channels 2 and 4 during 10 weeks post-op. Distinct changes in the neural signals were observed, more so after subtraction of the reference signal (channel 1) from the neural recordings (channels 2 and 4). This subtraction enhanced the view of the stimulation spikes.
CONCLUSIONS: These preliminary results indicated that ONI could deliver electrical stimulation that could be recorded by the proximal electrode (afferent CNAPs), a critical first step in demonstrating the potential of this system for use in sensory prosthetic control. While encouraging, especially given the limited number of trials, additional research is necessary to confirm consistency across more extensive trials and subjects. The capability of the system to detect CNAPs may eventually facilitate the creation of prosthetics that provide a more organic sense of control by translating neural signals into active output in prosthetic limbs. This work is ongoing. Future work will focus on replicating these findings in the same animal over a period of 6 months, refining stimulation parameters, and exploring the integration of sensory feedback to enhance prosthetic functionality.