A strategy for labeling data for the neural adaptation of a powered lower limb prosthesis

John A. Spanias; Eric J. Perreault; Levi J. Hargrove

doi:10.1109/EMBC.2014.6944276

A strategy for labeling data for the neural adaptation of a powered lower limb prosthesis

John A. Spanias, Eric J. Perreault, Levi J. Hargrove

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

10 Scopus citations

Abstract

Pattern recognition algorithms that use EMG signals have been proposed to help control powered lower limb prostheses. These algorithms do not automatically compensate for disturbances in EMG signals, resulting in deterioration of algorithm accuracies. Supervised adaptive pattern recognition algorithms can solve this problem, but require correct labeling of new data. Information from embedded mechanical sensors can be compared to the characteristic gait profiles of the different modes to identify the mode of the user's most recent stride and provide a label for new data. The purpose of this study was to develop a gait pattern estimator (GPE) that could automatically make such a comparison. The GPE output was used to supervise an adaptive EMG-based pattern recognition algorithm. Our results indicate that using GPE-based adaptation helped prevent classification errors that would otherwise occur between experimental sessions. The GPE could accurately label new data with a low error rate of approx. 2%. The low error rate of the GPE was reflected in the accuracy of an adapted pattern recognition algorithm. The error rate of the adapted algorithm that was supervised by the GPE was not significantly different from one that used perfect supervision.

Original language	English (US)
Title of host publication	2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	3090-3093
Number of pages	4
ISBN (Electronic)	9781424479290
DOIs	https://doi.org/10.1109/EMBC.2014.6944276
State	Published - Nov 2 2014
Event	2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014 - Chicago, United States Duration: Aug 26 2014 → Aug 30 2014

Publication series

Name	2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014

Other

Other	2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014
Country/Territory	United States
City	Chicago
Period	8/26/14 → 8/30/14

ASJC Scopus subject areas

Health Informatics
Computer Science Applications
Biomedical Engineering

Access to Document

10.1109/EMBC.2014.6944276

Cite this

Spanias, J. A., Perreault, E. J., & Hargrove, L. J. (2014). A strategy for labeling data for the neural adaptation of a powered lower limb prosthesis. In 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014 (pp. 3090-3093). Article 6944276 (2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/EMBC.2014.6944276

Spanias, John A. ; Perreault, Eric J. ; Hargrove, Levi J. / A strategy for labeling data for the neural adaptation of a powered lower limb prosthesis. 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 3090-3093 (2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014).

@inproceedings{1ed599ae99ce47cd87926a569a205bd1,

title = "A strategy for labeling data for the neural adaptation of a powered lower limb prosthesis",

abstract = "Pattern recognition algorithms that use EMG signals have been proposed to help control powered lower limb prostheses. These algorithms do not automatically compensate for disturbances in EMG signals, resulting in deterioration of algorithm accuracies. Supervised adaptive pattern recognition algorithms can solve this problem, but require correct labeling of new data. Information from embedded mechanical sensors can be compared to the characteristic gait profiles of the different modes to identify the mode of the user's most recent stride and provide a label for new data. The purpose of this study was to develop a gait pattern estimator (GPE) that could automatically make such a comparison. The GPE output was used to supervise an adaptive EMG-based pattern recognition algorithm. Our results indicate that using GPE-based adaptation helped prevent classification errors that would otherwise occur between experimental sessions. The GPE could accurately label new data with a low error rate of approx. 2%. The low error rate of the GPE was reflected in the accuracy of an adapted pattern recognition algorithm. The error rate of the adapted algorithm that was supervised by the GPE was not significantly different from one that used perfect supervision.",

author = "Spanias, {John A.} and Perreault, {Eric J.} and Hargrove, {Levi J.}",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014 ; Conference date: 26-08-2014 Through 30-08-2014",

year = "2014",

month = nov,

day = "2",

doi = "10.1109/EMBC.2014.6944276",

language = "English (US)",

series = "2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "3090--3093",

booktitle = "2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014",

address = "United States",

}

Spanias, JA, Perreault, EJ & Hargrove, LJ 2014, A strategy for labeling data for the neural adaptation of a powered lower limb prosthesis. in 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014., 6944276, 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014, Institute of Electrical and Electronics Engineers Inc., pp. 3090-3093, 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014, Chicago, United States, 8/26/14. https://doi.org/10.1109/EMBC.2014.6944276

A strategy for labeling data for the neural adaptation of a powered lower limb prosthesis. / Spanias, John A.; Perreault, Eric J.; Hargrove, Levi J.
2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 3090-3093 6944276 (2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - A strategy for labeling data for the neural adaptation of a powered lower limb prosthesis

AU - Spanias, John A.

AU - Perreault, Eric J.

AU - Hargrove, Levi J.

PY - 2014/11/2

Y1 - 2014/11/2

N2 - Pattern recognition algorithms that use EMG signals have been proposed to help control powered lower limb prostheses. These algorithms do not automatically compensate for disturbances in EMG signals, resulting in deterioration of algorithm accuracies. Supervised adaptive pattern recognition algorithms can solve this problem, but require correct labeling of new data. Information from embedded mechanical sensors can be compared to the characteristic gait profiles of the different modes to identify the mode of the user's most recent stride and provide a label for new data. The purpose of this study was to develop a gait pattern estimator (GPE) that could automatically make such a comparison. The GPE output was used to supervise an adaptive EMG-based pattern recognition algorithm. Our results indicate that using GPE-based adaptation helped prevent classification errors that would otherwise occur between experimental sessions. The GPE could accurately label new data with a low error rate of approx. 2%. The low error rate of the GPE was reflected in the accuracy of an adapted pattern recognition algorithm. The error rate of the adapted algorithm that was supervised by the GPE was not significantly different from one that used perfect supervision.

AB - Pattern recognition algorithms that use EMG signals have been proposed to help control powered lower limb prostheses. These algorithms do not automatically compensate for disturbances in EMG signals, resulting in deterioration of algorithm accuracies. Supervised adaptive pattern recognition algorithms can solve this problem, but require correct labeling of new data. Information from embedded mechanical sensors can be compared to the characteristic gait profiles of the different modes to identify the mode of the user's most recent stride and provide a label for new data. The purpose of this study was to develop a gait pattern estimator (GPE) that could automatically make such a comparison. The GPE output was used to supervise an adaptive EMG-based pattern recognition algorithm. Our results indicate that using GPE-based adaptation helped prevent classification errors that would otherwise occur between experimental sessions. The GPE could accurately label new data with a low error rate of approx. 2%. The low error rate of the GPE was reflected in the accuracy of an adapted pattern recognition algorithm. The error rate of the adapted algorithm that was supervised by the GPE was not significantly different from one that used perfect supervision.

UR - http://www.scopus.com/inward/record.url?scp=84929485144&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84929485144&partnerID=8YFLogxK

U2 - 10.1109/EMBC.2014.6944276

DO - 10.1109/EMBC.2014.6944276

M3 - Conference contribution

C2 - 25570644

AN - SCOPUS:84929485144

T3 - 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014

SP - 3090

EP - 3093

BT - 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014

Y2 - 26 August 2014 through 30 August 2014

ER -

Spanias JA, Perreault EJ , Hargrove LJ. A strategy for labeling data for the neural adaptation of a powered lower limb prosthesis. In 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014. Institute of Electrical and Electronics Engineers Inc. 2014. p. 3090-3093. 6944276. (2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2014). doi: 10.1109/EMBC.2014.6944276

A strategy for labeling data for the neural adaptation of a powered lower limb prosthesis

Abstract

Publication series

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this