Using latent representations of muscle activation patterns to mitigate myoelectric interface noise

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

Myoelectric controllers for upper limb prostheses are susceptible to signal disturbances across practical conditions. In particular, electrode liftoff or wire breakage introduce interface noise that, even if only present in a single channel, is detrimental to controller performance. We trained a supervised denoising variational autoencoder to learn a low-dimensional subspace underlying muscle activation patterns that was robust to noise in single EMG channels. Two latent space classifiers, which used the deep learning model, and two conventional LDA-based classifiers were used to classify wrist and hand gestures from clean and synthetically corrupted EMG signals. The baseline LDA classifier, trained on clean data only, suffered a marked increase in errors when evaluated on the corrupted data. The second LDA classifier, trained on clean and corrupted data, improved robustness to noise. Regardless, both latent space methods significantly outperformed both LDA methods in classifying clean and corrupted data. These results highlight that interface noise has adverse effects on current pattern recognition controllers but that deep learning inspired latent space classifiers can mitigate these effects and achieve highly accurate movement classification.

Original languageEnglish (US)
Title of host publication2021 10th International IEEE/EMBS Conference on Neural Engineering, NER 2021
PublisherIEEE Computer Society
Pages1148-1151
Number of pages4
ISBN (Electronic)9781728143378
DOIs
StatePublished - May 4 2021
Event10th International IEEE/EMBS Conference on Neural Engineering, NER 2021 - Virtual, Online, Italy
Duration: May 4 2021May 6 2021

Publication series

NameInternational IEEE/EMBS Conference on Neural Engineering, NER
Volume2021-May
ISSN (Print)1948-3546
ISSN (Electronic)1948-3554

Conference

Conference10th International IEEE/EMBS Conference on Neural Engineering, NER 2021
Country/TerritoryItaly
CityVirtual, Online
Period5/4/215/6/21

Keywords

  • Latent representation
  • Myoelectric control
  • Neural networks
  • Pattern recognition
  • Prosthesis

ASJC Scopus subject areas

  • Artificial Intelligence
  • Mechanical Engineering

Fingerprint

Dive into the research topics of 'Using latent representations of muscle activation patterns to mitigate myoelectric interface noise'. Together they form a unique fingerprint.

Cite this