Seamless and Intuitive Control of a Powered Prosthetic Leg Using Deep Neural Network for Transfemoral Amputees

Overview

Journal Wearable Technol

Publisher Cambridge University Press

Date 2023 Apr 12

PMID 37041885

Authors

Minjae Kim

Ann M Simon

Levi J Hargrove

Affiliations

Soon will be listed here.

Abstract

Powered prosthetic legs are becoming a promising option for amputee patients. However, developing safe, robust, and intuitive control strategies for powered legs remains one of the greatest challenges. Although a variety of control strategies have been proposed, creating and fine-tuning the system parameters is time-intensive and complicated when more activities need to be restored. In this study, we developed a deep neural network (DNN) model that facilitates seamless and intuitive gait generation and transitions across five ambulation modes: level-ground walking, ascending/descending ramps, and ascending/descending stairs. The combination of latent and time sequence features generated the desired impedance parameters within the ambulation modes and allowed seamless transitions between ambulation modes. The model was applied to the open-source bionic leg and tested on unilateral transfemoral users. It achieved the overall coefficient of determination of 0.72 with the state machine-based impedance parameters in the offline testing session. In addition, users were able to perform in-laboratory ambulation modes with an overall success rate of 96% during the online testing session. The results indicate that the DNN model is a promising candidate for subject-independent and tuning-free prosthetic leg control for transfemoral amputees.

Citing Articles

A Deep Learning Framework for End-to-End Control of Powered Prostheses.

Nuesslein C, Young A IEEE Robot Autom Lett. 2025; 9(5):3988-3994.

PMID: 40012860 PMC: 11864809. DOI: 10.1109/lra.2024.3374189.

Controlling Powered Prosthesis Kinematics Over Continuous Inter-Leg Transitions Between Walking and Stair Ascent/Descent.

Cheng S, Laubscher C, Gregg R IEEE Trans Neural Syst Rehabil Eng. 2024; 32:3891-3901.

PMID: 39446547 PMC: 11608573. DOI: 10.1109/TNSRE.2024.3485643.

Data-efficient human walking speed intent identification.

Higgins T, Bresingham K, Schmiedeler J, Wensing P Wearable Technol. 2024; 4:e19.

PMID: 38487770 PMC: 10936302. DOI: 10.1017/wtc.2023.15.

Controlling Powered Prosthesis Kinematics over Continuous Transitions Between Walk and Stair Ascent.

Cheng S, Laubscher C, Gregg R Rep U S. 2023; 2023:2108-2115.

PMID: 38130335 PMC: 10732262. DOI: 10.1109/iros55552.2023.10341457.

Using Deep Learning Models to Predict Prosthetic Ankle Torque.

Prasanna C, Realmuto J, Anderson A, Rombokas E, Klute G Sensors (Basel). 2023; 23(18).

PMID: 37765769 PMC: 10535406. DOI: 10.3390/s23187712.

References

Laschowski B, McNally W, Wong A, McPhee J . Environment Classification for Robotic Leg Prostheses and Exoskeletons Using Deep Convolutional Neural Networks. Front Neurorobot. 2022; 15:730965. PMC: 8855111. DOI: 10.3389/fnbot.2021.730965. View

Graupe D, Salahi J, Kohn K . Multifunctional prosthesis and orthosis control via microcomputer identification of temporal pattern differences in single-site myoelectric signals. J Biomed Eng. 1982; 4(1):17-22. DOI: 10.1016/0141-5425(82)90021-8. View

Montgomery J, Grabowski A . Use of a powered ankle-foot prosthesis reduces the metabolic cost of uphill walking and improves leg work symmetry in people with transtibial amputations. J R Soc Interface. 2018; 15(145). PMC: 6127176. DOI: 10.1098/rsif.2018.0442. View

Kim J, Wensman J, Colabianchi N, Gates D . The influence of powered prostheses on user perspectives, metabolics, and activity: a randomized crossover trial. J Neuroeng Rehabil. 2021; 18(1):49. PMC: 7962267. DOI: 10.1186/s12984-021-00842-2. View

Huang H, Zhang F, Hargrove L, Dou Z, Rogers D, Englehart K . Continuous locomotion-mode identification for prosthetic legs based on neuromuscular-mechanical fusion. IEEE Trans Biomed Eng. 2011; 58(10):2867-75. PMC: 3235670. DOI: 10.1109/TBME.2011.2161671. View

Quintero D, Villarreal D, Lambert D, Kapp S, Gregg R . Continuous-Phase Control of a Powered Knee-Ankle Prosthesis: Amputee Experiments Across Speeds and Inclines. IEEE Trans Robot. 2018; 34(3):686-701. PMC: 6042879. DOI: 10.1109/TRO.2018.2794536. View

Villarreal D, Gregg R . Unified Phase Variables of Relative Degree Two for Human Locomotion. Annu Int Conf IEEE Eng Med Biol Soc. 2017; 2016:6262-6267. PMC: 5324719. DOI: 10.1109/EMBC.2016.7592160. View

Sup F, Bohara A, Goldfarb M . Design and Control of a Powered Transfemoral Prosthesis. Int J Rob Res. 2009; 27(2):263-273. PMC: 2773553. DOI: 10.1177/0278364907084588. View

Huang H, Crouch D, Liu M, Sawicki G, Wang D . A Cyber Expert System for Auto-Tuning Powered Prosthesis Impedance Control Parameters. Ann Biomed Eng. 2015; 44(5):1613-24. DOI: 10.1007/s10439-015-1464-7. View

10.

Azocar A, Mooney L, Duval J, Simon A, Hargrove L, Rouse E . Design and clinical implementation of an open-source bionic leg. Nat Biomed Eng. 2020; 4(10):941-953. PMC: 7581510. DOI: 10.1038/s41551-020-00619-3. View

11.

Rabe K, Jahanandish M, Boehm J, Fey A, Hoyt K, Fey N . Ultrasound Sensing Can Improve Continuous Classification of Discrete Ambulation Modes Compared to Surface Electromyography. IEEE Trans Biomed Eng. 2020; 68(4):1379-1388. DOI: 10.1109/TBME.2020.3032077. View

12.

Hudgins B, Parker P, Scott R . A new strategy for multifunction myoelectric control. IEEE Trans Biomed Eng. 1993; 40(1):82-94. DOI: 10.1109/10.204774. View

13.

Spanias J, Simon A, Finucane S, Perreault E, Hargrove L . Online adaptive neural control of a robotic lower limb prosthesis. J Neural Eng. 2017; 15(1):016015. PMC: 5802866. DOI: 10.1088/1741-2552/aa92a8. View

14.

Hu B, Simon A, Hargrove L . Deep generative models with data augmentation to learn robust representations of movement intention for powered leg prostheses. IEEE Trans Med Robot Bionics. 2022; 1(4):267-278. PMC: 9499185. DOI: 10.1109/tmrb.2019.2952148. View

15.

Au S, Berniker M, Herr H . Powered ankle-foot prosthesis to assist level-ground and stair-descent gaits. Neural Netw. 2008; 21(4):654-66. DOI: 10.1016/j.neunet.2008.03.006. View

16.

Quintero D, Lambert D, Villarreal D, Gregg R . Real-Time Continuous Gait Phase and Speed Estimation from a Single Sensor. Control Technol Appl. 2018; 2017:847-852. PMC: 6105289. DOI: 10.1109/CCTA.2017.8062565. View

17.

Kang I, Molinaro D, Duggal S, Chen Y, Kunapuli P, Young A . Real-Time Gait Phase Estimation for Robotic Hip Exoskeleton Control During Multimodal Locomotion. IEEE Robot Autom Lett. 2021; 6(2):3491-3497. PMC: 8488948. DOI: 10.1109/lra.2021.3062562. View

18.

Liu M, Zhang F, Huang H . An Adaptive Classification Strategy for Reliable Locomotion Mode Recognition. Sensors (Basel). 2017; 17(9). PMC: 5621085. DOI: 10.3390/s17092020. View

19.

Simon A, Ingraham K, Fey N, Finucane S, Lipschutz R, Young A . Configuring a powered knee and ankle prosthesis for transfemoral amputees within five specific ambulation modes. PLoS One. 2014; 9(6):e99387. PMC: 4051756. DOI: 10.1371/journal.pone.0099387. View

20.

Quintero D, Villarreal D, Gregg R . Preliminary Experiments with a Unified Controller for a Powered Knee-Ankle Prosthetic Leg Across Walking Speeds. Rep U S. 2017; 2016:5427-5433. PMC: 5381823. DOI: 10.1109/IROS.2016.7759798. View