Intent recognition in a powered lower limb prosthesis using time history information

Aaron J. Young, Ann M. Simon, Nicholas P. Fey, Levi J. Hargrove

Research output: Contribution to journalArticle

56 Citations (Scopus)

Abstract

New computerized and powered lower limb pros-theses are being developed that enable amputees to perform multiple locomotion modes. However, current lower limb prosthesis controllers are not capable of transitioning these devices automatically and seamlessly between locomotion modes such as level-ground walking, stairs and slopes. The focus of this study was to evaluate different intent recognition interfaces, which if configured properly, may be capable of providing more natural transitions between locomotion modes. Intent recognition can be accomplished using a multitude of different signals from mechanical sensors on the prosthesis. Since these signals are non-stationary over any given stride, and gait is cyclical, time history information may improve locomotion mode recognition. The authors propose a dynamic Bayesian network classification strategy to incorporate prior sensor information over the gait cycle with current sensor information. Six transfemoral amputees performed locomotion circuits comprising level-ground walking and ascending/descending stairs and ramps using a powered knee and ankle prosthesis. Using time history reduced steady-state misclassifications by over half (p<0.01), when compared to strategies that did not use time history, without reducing intent recognition performance during transitions. These results suggest that including time history information across the gait cycle can enhance locomotion mode intent recognition performance.

Original languageEnglish (US)
Pages (from-to)631-641
Number of pages11
JournalAnnals of Biomedical Engineering
Volume42
Issue number3
DOIs
StatePublished - Jan 1 2014

Fingerprint

Stairs
Sensors
Bayesian networks
Controllers
Networks (circuits)
Prostheses and Implants

Keywords

  • Dynamic Bayesian network
  • Intent recognition
  • Non-stationary signal analysis
  • Powered lower limb prosthesis
  • Transfemoral amputee

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Intent recognition in a powered lower limb prosthesis using time history information. / Young, Aaron J.; Simon, Ann M.; Fey, Nicholas P.; Hargrove, Levi J.

In: Annals of Biomedical Engineering, Vol. 42, No. 3, 01.01.2014, p. 631-641.

Research output: Contribution to journalArticle

Young, Aaron J. ; Simon, Ann M. ; Fey, Nicholas P. ; Hargrove, Levi J. / Intent recognition in a powered lower limb prosthesis using time history information. In: Annals of Biomedical Engineering. 2014 ; Vol. 42, No. 3. pp. 631-641.
@article{5ff21d3370de44dc820199804dcae8ea,
title = "Intent recognition in a powered lower limb prosthesis using time history information",
abstract = "New computerized and powered lower limb pros-theses are being developed that enable amputees to perform multiple locomotion modes. However, current lower limb prosthesis controllers are not capable of transitioning these devices automatically and seamlessly between locomotion modes such as level-ground walking, stairs and slopes. The focus of this study was to evaluate different intent recognition interfaces, which if configured properly, may be capable of providing more natural transitions between locomotion modes. Intent recognition can be accomplished using a multitude of different signals from mechanical sensors on the prosthesis. Since these signals are non-stationary over any given stride, and gait is cyclical, time history information may improve locomotion mode recognition. The authors propose a dynamic Bayesian network classification strategy to incorporate prior sensor information over the gait cycle with current sensor information. Six transfemoral amputees performed locomotion circuits comprising level-ground walking and ascending/descending stairs and ramps using a powered knee and ankle prosthesis. Using time history reduced steady-state misclassifications by over half (p<0.01), when compared to strategies that did not use time history, without reducing intent recognition performance during transitions. These results suggest that including time history information across the gait cycle can enhance locomotion mode intent recognition performance.",
keywords = "Dynamic Bayesian network, Intent recognition, Non-stationary signal analysis, Powered lower limb prosthesis, Transfemoral amputee",
author = "Young, {Aaron J.} and Simon, {Ann M.} and Fey, {Nicholas P.} and Hargrove, {Levi J.}",
year = "2014",
month = "1",
day = "1",
doi = "10.1007/s10439-013-0909-0",
language = "English (US)",
volume = "42",
pages = "631--641",
journal = "Annals of Biomedical Engineering",
issn = "0090-6964",
publisher = "Springer Netherlands",
number = "3",

}

TY - JOUR

T1 - Intent recognition in a powered lower limb prosthesis using time history information

AU - Young, Aaron J.

AU - Simon, Ann M.

AU - Fey, Nicholas P.

AU - Hargrove, Levi J.

PY - 2014/1/1

Y1 - 2014/1/1

N2 - New computerized and powered lower limb pros-theses are being developed that enable amputees to perform multiple locomotion modes. However, current lower limb prosthesis controllers are not capable of transitioning these devices automatically and seamlessly between locomotion modes such as level-ground walking, stairs and slopes. The focus of this study was to evaluate different intent recognition interfaces, which if configured properly, may be capable of providing more natural transitions between locomotion modes. Intent recognition can be accomplished using a multitude of different signals from mechanical sensors on the prosthesis. Since these signals are non-stationary over any given stride, and gait is cyclical, time history information may improve locomotion mode recognition. The authors propose a dynamic Bayesian network classification strategy to incorporate prior sensor information over the gait cycle with current sensor information. Six transfemoral amputees performed locomotion circuits comprising level-ground walking and ascending/descending stairs and ramps using a powered knee and ankle prosthesis. Using time history reduced steady-state misclassifications by over half (p<0.01), when compared to strategies that did not use time history, without reducing intent recognition performance during transitions. These results suggest that including time history information across the gait cycle can enhance locomotion mode intent recognition performance.

AB - New computerized and powered lower limb pros-theses are being developed that enable amputees to perform multiple locomotion modes. However, current lower limb prosthesis controllers are not capable of transitioning these devices automatically and seamlessly between locomotion modes such as level-ground walking, stairs and slopes. The focus of this study was to evaluate different intent recognition interfaces, which if configured properly, may be capable of providing more natural transitions between locomotion modes. Intent recognition can be accomplished using a multitude of different signals from mechanical sensors on the prosthesis. Since these signals are non-stationary over any given stride, and gait is cyclical, time history information may improve locomotion mode recognition. The authors propose a dynamic Bayesian network classification strategy to incorporate prior sensor information over the gait cycle with current sensor information. Six transfemoral amputees performed locomotion circuits comprising level-ground walking and ascending/descending stairs and ramps using a powered knee and ankle prosthesis. Using time history reduced steady-state misclassifications by over half (p<0.01), when compared to strategies that did not use time history, without reducing intent recognition performance during transitions. These results suggest that including time history information across the gait cycle can enhance locomotion mode intent recognition performance.

KW - Dynamic Bayesian network

KW - Intent recognition

KW - Non-stationary signal analysis

KW - Powered lower limb prosthesis

KW - Transfemoral amputee

UR - http://www.scopus.com/inward/record.url?scp=84898663099&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84898663099&partnerID=8YFLogxK

U2 - 10.1007/s10439-013-0909-0

DO - 10.1007/s10439-013-0909-0

M3 - Article

C2 - 24052324

AN - SCOPUS:84898663099

VL - 42

SP - 631

EP - 641

JO - Annals of Biomedical Engineering

JF - Annals of Biomedical Engineering

SN - 0090-6964

IS - 3

ER -