Finger Force Decoding from Motor Units Activity on Neuromorphic Hardware

TL;DR

This paper introduces a neuromorphic hardware-based method for real-time finger force estimation using motor neuron spike trains from EMG, achieving accurate predictions with low power suitable for embedded neurotechnology applications.

Contribution

It presents the first motor neuron-based continuous regression approach on neuromorphic hardware for finger force decoding, reducing power consumption and computational requirements.

Findings

Accurate finger force prediction demonstrated on neuromorphic hardware.

Low-power, real-time inference achieved with spike train-based signals.

First demonstration of motor neuron-based regression on neuromorphic systems.

Abstract

Accurate finger force estimation is critical for next-generation human-machine interfaces. Traditional electromyography (EMG)-based decoding methods using deep learning require large datasets and high computational resources, limiting their use in real-time, embedded systems. Here, we propose a novel approach that performs finger force regression using spike trains from individual motor neurons, extracted from high-density EMG. These biologically grounded signals drive a spiking neural network implemented on a mixed-signal neuromorphic processor. Unlike prior work that encodes EMG into events, our method exploits spike timing on motor units to perform low-power, real-time inference. This is the first demonstration of motor neuron-based continuous regression computed directly on neuromorphic hardware. Our results confirm accurate finger-specific force prediction with minimal energy use, opening new possibilities for embedded decoding in prosthetics and wearable neurotechnology.