Development of a deep neural network for automated electromyographic pattern classification

Abstract

Determining the signal quality of surface electromyography (sEMG) recordings is time consuming and requires the judgment of trained observers. An automated procedure to evaluate sEMG quality would streamline data processing and reduce time demands. This paper compares the performance of two supervised and three unsupervised artificial neural networks (ANNs) in evaluation of sEMG quality. Manually classified sEMG recordings from various lower-limb muscles during motor tasks were used to train (n=28000), test performance (n=12000), and evaluate accuracy (n=47000) of the five ANNs in classifying signals into four categories. Unsupervised ANNs demonstrated a 30-40% increase in classification accuracy (>98%) compared to supervised ANNs. AlexNet demonstrated the highest accuracy (99.55%) with negligible false classifications. Results indicate that sEMG quality evaluation can be automated via an ANN without compromising human-like classification accuracy. This classifier will be publicly available and will be a valuable tool for researchers and clinicians using electromyography.