Classification of naturally evoked compound action potentials in peripheral nerve spatiotemporal recordings

Ryan G L Koh,Adrian I Nachman,José Zariffa

doi:10.1038/s41598-019-47450-8

Abstract

Peripheral neural signals have the potential to provide the necessary motor, sensory or autonomic information for robust control in many neuroprosthetic and neuromodulation applications. However, developing methods to recover information encoded in these signals is a significant challenge. We introduce the idea of using spatiotemporal signatures extracted from multi-contact nerve cuff electrode recordings to classify naturally evoked compound action potentials (CAP). 9 Long-Evan rats were implanted with a 56-channel nerve cuff on the sciatic nerve. Afferent activity was selectively evoked in the different fascicles of the sciatic nerve (tibial, peroneal, sural) using mechano-sensory stimuli. Spatiotemporal signatures of recorded CAPs were used to train three different classifiers. Performance was measured based on the classification accuracy, F1-score, and the ability to reconstruct original firing rates of neural pathways. The mean classification accuracies, for a 3-class problem, for the best performing classifier was 0.686 ± 0.126 and corresponding mean F1-score was 0.605 ± 0.212. The mean Pearson correlation coefficients between the original firing rates and estimated firing rates found for the best classifier was 0.728 ± 0.276. The proposed method demonstrates the possibility of classifying individual naturally evoked CAPs in peripheral neural signals recorded from extraneural electrodes, allowing for more precise control signals in neuroprosthetic applications.

Highlights

The development of neural interfaces for recording peripheral nerve activity, including signal processing algorithms to analyze this activity, is a rapidly growing area of research[1,2,3,4,5,6,7]
Of interest here is the task of selective recording, which refers to the ability of a neural interface to discriminate between the activity of different neural pathways
The classification accuracies and firing rate reconstructions achieved in this study support the feasibility of compound action potentials (CAP)-based classification from multi-contact nerve cuff electrodes

Summary

Introduction

The development of neural interfaces for recording peripheral nerve activity, including signal processing algorithms to analyze this activity, is a rapidly growing area of research[1,2,3,4,5,6,7]. Peripheral nerve interfaces to date have had limited success in implementing selective recording strategies that are suitable for long-term use in humans. Intraneural electrodes that penetrate the nerve, such as longitudinal or transverse intrafascicular electrodes or microelectrode arrays, are able to selectively record neural activity and are increasingly being translated to human studies[10,19,20,21,22,23] but still struggle to demonstrate viability for stable chronic www.nature.com/scientificreports/. Extraneural electrodes, such as nerve cuffs[24,25,26] or flat interface nerve electrodes (FINEs)[27,28], have been shown to be stable for chronic implantation in humans for recording and stimulation applications. Nerve cuff electrodes are an appealing choice of neural interface, but it is a challenge to achieve sufficient recording selectivity with these devices

Methods

Results

Conclusion