Across-Day Lower Limb Pattern Recognition Performance of a Powered Knee-Ankle Prosthesis

Annie Simon*, Emily A. Seyforth, Levi J Hargrove

*Corresponding author for this work

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

Abstract

Powered lower limb prostheses have the capabilities to assist individuals with a lower limb amputation during ambulation. While these devices can generate power at the knee and/or ankle to assist with incline walking and stair climbing, it is difficult to control the transition between these ambulation modes in a seamless and natural way. Pattern recognition has been suggested as an alternative to using a key fob to switch between modes and recent results have shown reliable performance (less than 5% error rate) across five ambulation modes. In this study we investigated performance of a similar system across multiple sessions of use, a necessary step prior to clinical use. Two individuals with a transfemoral amputation used a powered knee-ankle for five ambulation activities including level-ground walking, ramp ascent, ramp descent, stair ascent, and stair descent over four sessions spaced out over at least two months. An intent recognition system was trained using embedded prosthesis mechanical sensors with varying amounts of data collected across the sessions to determine the effect of multi-session use and increased variation in the activities trained. Overall system error rate decreased from 1.45% [0.3%] when the system was trained with Session 1 data only and tested with Session 4 data to 0.60% [0.02%] when the system was trained with Sessions 1-3 data and tested with Session 4 data. These results demonstrate that a reliable intent recognition system can be created with multiple sessions of use, bringing lower limb intent recognition systems for powered prostheses one step closer to clinical viability.

Original languageEnglish (US)
Title of host publicationBIOROB 2018 - 7th IEEE International Conference on Biomedical Robotics and Biomechatronics
PublisherIEEE Computer Society
Pages242-247
Number of pages6
ISBN (Electronic)9781538681831
DOIs
StatePublished - Oct 9 2018
Event7th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, BIOROB 2018 - Enschede, Netherlands
Duration: Aug 26 2018Aug 29 2018

Publication series

NameProceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics
Volume2018-August
ISSN (Print)2155-1774

Other

Other7th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, BIOROB 2018
CountryNetherlands
CityEnschede
Period8/26/188/29/18

Fingerprint

Knee prostheses
Stairs
Pattern recognition
Prosthetics
Switches
Sensors

ASJC Scopus subject areas

  • Artificial Intelligence
  • Biomedical Engineering
  • Mechanical Engineering

Cite this

Simon, A., Seyforth, E. A., & Hargrove, L. J. (2018). Across-Day Lower Limb Pattern Recognition Performance of a Powered Knee-Ankle Prosthesis. In BIOROB 2018 - 7th IEEE International Conference on Biomedical Robotics and Biomechatronics (pp. 242-247). [8487836] (Proceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics; Vol. 2018-August). IEEE Computer Society. https://doi.org/10.1109/BIOROB.2018.8487836
Simon, Annie ; Seyforth, Emily A. ; Hargrove, Levi J. / Across-Day Lower Limb Pattern Recognition Performance of a Powered Knee-Ankle Prosthesis. BIOROB 2018 - 7th IEEE International Conference on Biomedical Robotics and Biomechatronics. IEEE Computer Society, 2018. pp. 242-247 (Proceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics).
@inproceedings{c8e38c446518450999dd2c948ef1ef22,
title = "Across-Day Lower Limb Pattern Recognition Performance of a Powered Knee-Ankle Prosthesis",
abstract = "Powered lower limb prostheses have the capabilities to assist individuals with a lower limb amputation during ambulation. While these devices can generate power at the knee and/or ankle to assist with incline walking and stair climbing, it is difficult to control the transition between these ambulation modes in a seamless and natural way. Pattern recognition has been suggested as an alternative to using a key fob to switch between modes and recent results have shown reliable performance (less than 5{\%} error rate) across five ambulation modes. In this study we investigated performance of a similar system across multiple sessions of use, a necessary step prior to clinical use. Two individuals with a transfemoral amputation used a powered knee-ankle for five ambulation activities including level-ground walking, ramp ascent, ramp descent, stair ascent, and stair descent over four sessions spaced out over at least two months. An intent recognition system was trained using embedded prosthesis mechanical sensors with varying amounts of data collected across the sessions to determine the effect of multi-session use and increased variation in the activities trained. Overall system error rate decreased from 1.45{\%} [0.3{\%}] when the system was trained with Session 1 data only and tested with Session 4 data to 0.60{\%} [0.02{\%}] when the system was trained with Sessions 1-3 data and tested with Session 4 data. These results demonstrate that a reliable intent recognition system can be created with multiple sessions of use, bringing lower limb intent recognition systems for powered prostheses one step closer to clinical viability.",
author = "Annie Simon and Seyforth, {Emily A.} and Hargrove, {Levi J}",
year = "2018",
month = "10",
day = "9",
doi = "10.1109/BIOROB.2018.8487836",
language = "English (US)",
series = "Proceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics",
publisher = "IEEE Computer Society",
pages = "242--247",
booktitle = "BIOROB 2018 - 7th IEEE International Conference on Biomedical Robotics and Biomechatronics",
address = "United States",

}

Simon, A, Seyforth, EA & Hargrove, LJ 2018, Across-Day Lower Limb Pattern Recognition Performance of a Powered Knee-Ankle Prosthesis. in BIOROB 2018 - 7th IEEE International Conference on Biomedical Robotics and Biomechatronics., 8487836, Proceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics, vol. 2018-August, IEEE Computer Society, pp. 242-247, 7th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, BIOROB 2018, Enschede, Netherlands, 8/26/18. https://doi.org/10.1109/BIOROB.2018.8487836

Across-Day Lower Limb Pattern Recognition Performance of a Powered Knee-Ankle Prosthesis. / Simon, Annie; Seyforth, Emily A.; Hargrove, Levi J.

BIOROB 2018 - 7th IEEE International Conference on Biomedical Robotics and Biomechatronics. IEEE Computer Society, 2018. p. 242-247 8487836 (Proceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics; Vol. 2018-August).

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

TY - GEN

T1 - Across-Day Lower Limb Pattern Recognition Performance of a Powered Knee-Ankle Prosthesis

AU - Simon, Annie

AU - Seyforth, Emily A.

AU - Hargrove, Levi J

PY - 2018/10/9

Y1 - 2018/10/9

N2 - Powered lower limb prostheses have the capabilities to assist individuals with a lower limb amputation during ambulation. While these devices can generate power at the knee and/or ankle to assist with incline walking and stair climbing, it is difficult to control the transition between these ambulation modes in a seamless and natural way. Pattern recognition has been suggested as an alternative to using a key fob to switch between modes and recent results have shown reliable performance (less than 5% error rate) across five ambulation modes. In this study we investigated performance of a similar system across multiple sessions of use, a necessary step prior to clinical use. Two individuals with a transfemoral amputation used a powered knee-ankle for five ambulation activities including level-ground walking, ramp ascent, ramp descent, stair ascent, and stair descent over four sessions spaced out over at least two months. An intent recognition system was trained using embedded prosthesis mechanical sensors with varying amounts of data collected across the sessions to determine the effect of multi-session use and increased variation in the activities trained. Overall system error rate decreased from 1.45% [0.3%] when the system was trained with Session 1 data only and tested with Session 4 data to 0.60% [0.02%] when the system was trained with Sessions 1-3 data and tested with Session 4 data. These results demonstrate that a reliable intent recognition system can be created with multiple sessions of use, bringing lower limb intent recognition systems for powered prostheses one step closer to clinical viability.

AB - Powered lower limb prostheses have the capabilities to assist individuals with a lower limb amputation during ambulation. While these devices can generate power at the knee and/or ankle to assist with incline walking and stair climbing, it is difficult to control the transition between these ambulation modes in a seamless and natural way. Pattern recognition has been suggested as an alternative to using a key fob to switch between modes and recent results have shown reliable performance (less than 5% error rate) across five ambulation modes. In this study we investigated performance of a similar system across multiple sessions of use, a necessary step prior to clinical use. Two individuals with a transfemoral amputation used a powered knee-ankle for five ambulation activities including level-ground walking, ramp ascent, ramp descent, stair ascent, and stair descent over four sessions spaced out over at least two months. An intent recognition system was trained using embedded prosthesis mechanical sensors with varying amounts of data collected across the sessions to determine the effect of multi-session use and increased variation in the activities trained. Overall system error rate decreased from 1.45% [0.3%] when the system was trained with Session 1 data only and tested with Session 4 data to 0.60% [0.02%] when the system was trained with Sessions 1-3 data and tested with Session 4 data. These results demonstrate that a reliable intent recognition system can be created with multiple sessions of use, bringing lower limb intent recognition systems for powered prostheses one step closer to clinical viability.

UR - http://www.scopus.com/inward/record.url?scp=85056606350&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85056606350&partnerID=8YFLogxK

U2 - 10.1109/BIOROB.2018.8487836

DO - 10.1109/BIOROB.2018.8487836

M3 - Conference contribution

T3 - Proceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics

SP - 242

EP - 247

BT - BIOROB 2018 - 7th IEEE International Conference on Biomedical Robotics and Biomechatronics

PB - IEEE Computer Society

ER -

Simon A, Seyforth EA, Hargrove LJ. Across-Day Lower Limb Pattern Recognition Performance of a Powered Knee-Ankle Prosthesis. In BIOROB 2018 - 7th IEEE International Conference on Biomedical Robotics and Biomechatronics. IEEE Computer Society. 2018. p. 242-247. 8487836. (Proceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics). https://doi.org/10.1109/BIOROB.2018.8487836