Covariate shift adaptation in EMG pattern recognition for prosthetic device control

Marina M C Vidovic, Liliana P. Paredes, Hwang Han-Jeong Hwang, Sebastian Amsüss, Jaspar Pahl, Janne M. Hahne, Bernhard Graimann, Dario Farina, Klaus Robert Müller

Research output: Contribution to journalArticle

Abstract

Ensuring robustness of myocontrol algorithms for prosthetic devices is an important challenge. Robustness needs to be maintained under nonstationarities, e.g. due to electrode shifts after donning and doffing, sweating, additional weight or varying arm positions. Such nonstationary behavior changes the signal distributions - a scenario often referred to as covariate shift. This circumstance causes a significant decrease in classification accuracy in daily life applications. Re-training is possible but it is time consuming since it requires a large number of trials. In this paper, we propose to adapt the EMG classifier by a small calibration set only, which is able to capture the relevant aspects of the nonstationarities, but requires re-training data of only very short duration. We tested this strategy on signals acquired across 5 days in able-bodied individuals. The results showed that an estimator that shrinks the training model parameters towards the calibration set parameters significantly increased the classifier performance across different testing days. Even when using only one trial per class as re-training data for each day, the classification accuracy remained > 92% over five days. These results indicate that the proposed methodology can be a practical means for improving robustness in pattern recognition methods for myocontrol.

Fingerprint

Prosthetics
Calibration
Pattern recognition
Classifiers
Equipment and Supplies
Sweating
Electrodes
Weights and Measures
Testing

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Covariate shift adaptation in EMG pattern recognition for prosthetic device control. / Vidovic, Marina M C; Paredes, Liliana P.; Han-Jeong Hwang, Hwang; Amsüss, Sebastian; Pahl, Jaspar; Hahne, Janne M.; Graimann, Bernhard; Farina, Dario; Müller, Klaus Robert.

In: Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference, Vol. 2014, 2014, p. 4370-4373.

Research output: Contribution to journalArticle

Vidovic, Marina M C ; Paredes, Liliana P. ; Han-Jeong Hwang, Hwang ; Amsüss, Sebastian ; Pahl, Jaspar ; Hahne, Janne M. ; Graimann, Bernhard ; Farina, Dario ; Müller, Klaus Robert. / Covariate shift adaptation in EMG pattern recognition for prosthetic device control. In: Conference proceedings : ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference. 2014 ; Vol. 2014. pp. 4370-4373.
@article{0977307c543c4196875edeb856dbc34c,
title = "Covariate shift adaptation in EMG pattern recognition for prosthetic device control",
abstract = "Ensuring robustness of myocontrol algorithms for prosthetic devices is an important challenge. Robustness needs to be maintained under nonstationarities, e.g. due to electrode shifts after donning and doffing, sweating, additional weight or varying arm positions. Such nonstationary behavior changes the signal distributions - a scenario often referred to as covariate shift. This circumstance causes a significant decrease in classification accuracy in daily life applications. Re-training is possible but it is time consuming since it requires a large number of trials. In this paper, we propose to adapt the EMG classifier by a small calibration set only, which is able to capture the relevant aspects of the nonstationarities, but requires re-training data of only very short duration. We tested this strategy on signals acquired across 5 days in able-bodied individuals. The results showed that an estimator that shrinks the training model parameters towards the calibration set parameters significantly increased the classifier performance across different testing days. Even when using only one trial per class as re-training data for each day, the classification accuracy remained > 92{\%} over five days. These results indicate that the proposed methodology can be a practical means for improving robustness in pattern recognition methods for myocontrol.",
author = "Vidovic, {Marina M C} and Paredes, {Liliana P.} and {Han-Jeong Hwang}, Hwang and Sebastian Ams{\"u}ss and Jaspar Pahl and Hahne, {Janne M.} and Bernhard Graimann and Dario Farina and M{\"u}ller, {Klaus Robert}",
year = "2014",
doi = "10.1109/EMBC.2014.6944592",
language = "English",
volume = "2014",
pages = "4370--4373",
journal = "The BMJ",
issn = "0730-6512",
publisher = "Kluwer Academic Publishers",

}

TY - JOUR

T1 - Covariate shift adaptation in EMG pattern recognition for prosthetic device control

AU - Vidovic, Marina M C

AU - Paredes, Liliana P.

AU - Han-Jeong Hwang, Hwang

AU - Amsüss, Sebastian

AU - Pahl, Jaspar

AU - Hahne, Janne M.

AU - Graimann, Bernhard

AU - Farina, Dario

AU - Müller, Klaus Robert

PY - 2014

Y1 - 2014

N2 - Ensuring robustness of myocontrol algorithms for prosthetic devices is an important challenge. Robustness needs to be maintained under nonstationarities, e.g. due to electrode shifts after donning and doffing, sweating, additional weight or varying arm positions. Such nonstationary behavior changes the signal distributions - a scenario often referred to as covariate shift. This circumstance causes a significant decrease in classification accuracy in daily life applications. Re-training is possible but it is time consuming since it requires a large number of trials. In this paper, we propose to adapt the EMG classifier by a small calibration set only, which is able to capture the relevant aspects of the nonstationarities, but requires re-training data of only very short duration. We tested this strategy on signals acquired across 5 days in able-bodied individuals. The results showed that an estimator that shrinks the training model parameters towards the calibration set parameters significantly increased the classifier performance across different testing days. Even when using only one trial per class as re-training data for each day, the classification accuracy remained > 92% over five days. These results indicate that the proposed methodology can be a practical means for improving robustness in pattern recognition methods for myocontrol.

AB - Ensuring robustness of myocontrol algorithms for prosthetic devices is an important challenge. Robustness needs to be maintained under nonstationarities, e.g. due to electrode shifts after donning and doffing, sweating, additional weight or varying arm positions. Such nonstationary behavior changes the signal distributions - a scenario often referred to as covariate shift. This circumstance causes a significant decrease in classification accuracy in daily life applications. Re-training is possible but it is time consuming since it requires a large number of trials. In this paper, we propose to adapt the EMG classifier by a small calibration set only, which is able to capture the relevant aspects of the nonstationarities, but requires re-training data of only very short duration. We tested this strategy on signals acquired across 5 days in able-bodied individuals. The results showed that an estimator that shrinks the training model parameters towards the calibration set parameters significantly increased the classifier performance across different testing days. Even when using only one trial per class as re-training data for each day, the classification accuracy remained > 92% over five days. These results indicate that the proposed methodology can be a practical means for improving robustness in pattern recognition methods for myocontrol.

UR - http://www.scopus.com/inward/record.url?scp=84942114131&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84942114131&partnerID=8YFLogxK

U2 - 10.1109/EMBC.2014.6944592

DO - 10.1109/EMBC.2014.6944592

M3 - Article

C2 - 25570960

VL - 2014

SP - 4370

EP - 4373

JO - The BMJ

JF - The BMJ

SN - 0730-6512

ER -