EMG feature evaluation for movement control of upper extremity prostheses

TLDR

EMG features have been studied for controlling myoelectric upper‑extremity prostheses. The study assessed movement‑class discrimination, robustness, and computational cost of various EMG features—integral, variance, zero crossings, Willison amplitude, v‑order, log detectors, and autoregressive parameters—using Davies‑Bouldin cluster separability and K‑nearest‑neighbor classification on data from an above‑elbow amputee’s residual biceps and triceps. The EMG Histogram feature was introduced and proved to be the most effective among the evaluated features.

Abstract

A variety of EMG features have been evaluated for control of myoelectric upper extremity prostheses. Movement class discrimination, robustness, and computational complexity of these features have been investigated for different time window sizes and noise levels. The measurements include novel application of the Davies-Bouldin index, a measure of cluster separability, and the K-nearest neighbor nonparametric classifier. The features evaluated are the integral of average value, the variance, the number of zero crossings, the Willison amplitude, the v-order and log detectors, and autoregressive model parameters. A new feature, the EMG Histogram, is introduced and shown to be the most effective of the group. The experiments were done on the data acquired from the residual biceps and triceps muscle of an above-elbow amputee.