Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Decoding of individual finger movements from surface EMG signals using vector autoregressive hierarchical hidden Markov models (VARHHMM)

Malešević, Nebojša ; Marković, Dimitrije ; Kanitz, Gunter ; Controzzi, Marco ; Cipriani, Christian and Antfolk, Christian LU (2017) 2017 International Conference on Rehabilitation Robotics, ICORR 2017 p.1518-1523
Abstract

In this paper we present a novel method for predicting individual fingers movements from surface electromyography (EMG). The method is intended for real-time dexterous control of a multifunctional prosthetic hand device. The EMG data was recorded using 16 single-ended channels positioned on the forearm of healthy participants. Synchronously with the EMG recording, the subjects performed consecutive finger movements based on the visual cues. Our algorithm could be described in following steps: extracting mean average value (MAV) of the EMG to be used as the feature for classification, piece-wise linear modeling of EMG feature dynamics, implementation of hierarchical hidden Markov models (HHMM) to capture transitions between linear... (More)

In this paper we present a novel method for predicting individual fingers movements from surface electromyography (EMG). The method is intended for real-time dexterous control of a multifunctional prosthetic hand device. The EMG data was recorded using 16 single-ended channels positioned on the forearm of healthy participants. Synchronously with the EMG recording, the subjects performed consecutive finger movements based on the visual cues. Our algorithm could be described in following steps: extracting mean average value (MAV) of the EMG to be used as the feature for classification, piece-wise linear modeling of EMG feature dynamics, implementation of hierarchical hidden Markov models (HHMM) to capture transitions between linear models, and implementation of Bayesian inference as the classifier. The performance of our classifier was evaluated against commonly used real-time classifiers. The results show that the current algorithm setup classifies EMG data similarly to the best among tested classifiers but with equal or less computational complexity.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
Bayesian Inference, Hidden Markov models (HMM), Myoelectric control, Vector Autoregressive (VAR) models
host publication
2017 International Conference on Rehabilitation Robotics, ICORR 2017
article number
8009463
pages
6 pages
publisher
IEEE Computer Society
conference name
2017 International Conference on Rehabilitation Robotics, ICORR 2017
conference location
London, United Kingdom
conference dates
2017-07-17 - 2017-07-20
external identifiers
  • pmid:28814035
  • scopus:85034824331
ISBN
9781538622964
DOI
10.1109/ICORR.2017.8009463
language
English
LU publication?
yes
id
50159994-b27e-44a4-ade5-8c6ea69b567c
date added to LUP
2017-12-14 11:55:12
date last changed
2024-07-08 07:33:39
@inproceedings{50159994-b27e-44a4-ade5-8c6ea69b567c,
  abstract     = {{<p>In this paper we present a novel method for predicting individual fingers movements from surface electromyography (EMG). The method is intended for real-time dexterous control of a multifunctional prosthetic hand device. The EMG data was recorded using 16 single-ended channels positioned on the forearm of healthy participants. Synchronously with the EMG recording, the subjects performed consecutive finger movements based on the visual cues. Our algorithm could be described in following steps: extracting mean average value (MAV) of the EMG to be used as the feature for classification, piece-wise linear modeling of EMG feature dynamics, implementation of hierarchical hidden Markov models (HHMM) to capture transitions between linear models, and implementation of Bayesian inference as the classifier. The performance of our classifier was evaluated against commonly used real-time classifiers. The results show that the current algorithm setup classifies EMG data similarly to the best among tested classifiers but with equal or less computational complexity.</p>}},
  author       = {{Malešević, Nebojša and Marković, Dimitrije and Kanitz, Gunter and Controzzi, Marco and Cipriani, Christian and Antfolk, Christian}},
  booktitle    = {{2017 International Conference on Rehabilitation Robotics, ICORR 2017}},
  isbn         = {{9781538622964}},
  keywords     = {{Bayesian Inference; Hidden Markov models (HMM); Myoelectric control; Vector Autoregressive (VAR) models}},
  language     = {{eng}},
  month        = {{08}},
  pages        = {{1518--1523}},
  publisher    = {{IEEE Computer Society}},
  title        = {{Decoding of individual finger movements from surface EMG signals using vector autoregressive hierarchical hidden Markov models (VARHHMM)}},
  url          = {{http://dx.doi.org/10.1109/ICORR.2017.8009463}},
  doi          = {{10.1109/ICORR.2017.8009463}},
  year         = {{2017}},
}