Objective: Extraction of information from the peripheral nervous system can provide control signals in neuroprosthetic applications. However, the ability to selectively record from different pathways within peripheral nerves is limited. We investigated the integration of spatial and temporal information for pathway discrimination in peripheral nerves using measurements from a multi-contact nerve cuff electrode. Approach: Spatiotemporal templates were established for different neural pathways of interest, and used to obtain tailored matched filters for each of these pathways. Simulated measurements of compound action potentials propagating through the nerve in different test cases were used to evaluate classification accuracy, percentage of missed spikes, and ability to reconstruct the original firing rates of the neural pathways. Main Results: The mean Pearson correlation coefficients between the original firing rates and estimated firing rates over all tests cases was found to be 0.832±0.161, 0.421±0.145, 0.481±0.340 for our algorithm, Bayesian spatial filters, and velocity selective recordings respectively. Significance: The proposed method shows that the spatiotemporal templates were able to provide more robust spike detection and reliable pathway discrimination than these existing algorithms.

Objective. Recording and stimulating from the peripheral nervous system are becoming important components in a new generation of bioelectronics systems. Although neurostimulation has seen a history of successful chronic applications in humans, peripheral nerve recording in humans chronically remains a challenge. Multi-contact nerve cuff electrode configurations have the potential to improve recording selectivity. We introduce the idea of using a convolutional neural network (CNN) to associate recordings of individual naturally evoked compound action potentials (CAPs) with neural pathways of interest, by exploiting the spatiotemporal patterns in multi-contact nerve cuff recordings. Approach. Nine Long-Evan rats were implanted with a 56-channel nerve cuff electrode on the sciatic nerve and afferent activity was selectively evoked in different fascicles (tibial, peroneal, sural) using mechanical stimuli. A recurrent neural network was then used to predict joint angles based on the predicted firing patterns from the CNN. Performance was measured based on the classification accuracy, F1-score and the ability to track the ankle joint angle. Main results. Classification accuracy and F1-score of the best CNN configuration were and 0.747  ±  0.114, respectively. The mean Pearson correlation coefficient between the manually measured ankle angle and the angle predicted from the estimated firing rate was Significance. The proposed method demonstrates that CAP-based classification can be achieved with high accuracy and can be used to track a physiological meaningful measure (e.g. joint angle). These results provide a promising direction for realizing more effective and intuitive neuroprosthetic systems.