Icon Legend

This session is not in your schedule.

This session is in your schedule. Click again to remove it.

Orthopedic and Rehabilitation Engineering

Orthopedic and Rehabilitation Engineering - Poster Session A

Poster K10 - Clinical and Biomechanical Comparison of Passive and Electromyographic Powered Ankle Prostheses

Thursday, October 24, 2024

10:00 AM - 11:00 AM EST

Location: Exhibit Hall E, F & G

Presenting Author(s)

Avery Murray

Undergraduate Researcher
Joint Biomedical Engineering Program - NCSU/UNC
Raleigh, North Carolina, United States

Co-Author(s)

JT

Josh Tacca, PhD

Postdoctoral researcher
Joint Biomedical Engineering Program - NCSU/UNC, United States
AM

Austin L. Mituniewicz, B.S. (he/him/his)

NSF Graduate Research Fellow
Joint Biomedical Engineering Program - NCSU/UNC
Carrboro, North Carolina, United States

Primary Investigator(s)

HH

He (Helen) Huang, PhD

Principal Investigator
Joint Biomedical Engineering Program - NCSU/UNC, United States

Introduction: Current passive ankle prostheses offer limited range of motion of the joint and cannot generate net positive mechanical work, impeding people with transtibial amputation’s (PwTA) mobility and stability. Newer autonomously controlled powered ankle prostheses can produce net mechanical work similar to the biological ankle, normalizing biomechanics and reducing compensation from the intact limb during pre-selected tasks such as walking or stair-climbing (Au et al., Neural Netw. 21, 2008). However, postural stability and balance tasks are difficult to pre-program because they require anticipatory and compensatory postural adjustments. Another new technology in powered prostheses restores direct neural command over the ankle through real-time electromyography (EMG). Through direct EMG (dEMG) control, PwTA exhibit improved postural control and reduced compensation from the intact limb in balance tasks, such as an expected perturbation (Fleming et al., Wearable Technol. 2, 2021, Fleming et al. Sci. Robot. 8, 2023). A powered ankle prosthesis with dEMG may also provide functional benefits in clinical tasks related to mobility, stability, and balance. This study evaluates the effectiveness of a dEMG powered ankle prosthesis in clinical tasks such as the stand and reach and timed-up-and-go (TUG), with and without cognitive dual tasks (Clemens et al., Clinical rehabilitation 32, 2018). Additional biomechanical tasks such as an expected perturbation and postural sway tasks will be evaluated to further analyze the effect of dEMG on PwTA’s postural stability.

Materials and

Methods: This project is part of an ongoing study recruiting PwTA in various clinical and biomechanical tasks and presents data from one participant. The powered ankle prosthesis used rotates in the sagittal plane through the actuation of two pneumatic muscle pairs. The posterior pair of muscles is controlled by residual gastrocnemius activation, enabling plantarflexion, while the anterior pair of muscles is controlled by the residual tibialis anterior activation, enabling dorsiflexion. All tests were performed with the participant’s own passive device and the powered prosthesis and conducted by a licensed physical therapist. Clinical tasks included: stand and reach, static balance, and TUG; with and without a dual task to access the cognitive load, either counting backwards by three or an auditory Stroop test. The participant repeated the TUG after each dual task. We analyzed the distance reached for the stand and reach, and average reaction time and performance of the cognitive tests while seated and while walking. After clinical testing was completed, reflective markers were placed on the participant for biomechanical testing on an instrumented dual-belt treadmill. Marker positions and ground reaction forces were recorded in Vicon and processed in Visual 3D. The participants performed an expected push and pull perturbation test. Postural sway tests were conducted in three different stances: 1) comfortable stance (shoulder width), 2) wide stance (10cm wider than shoulder), 3) crouched stance (~20 degree knee flexion). Center of pressure (CoP) excursion and ankle joint angles were measured and analyzed in the perturbation and postural sway tasks.

Results, Conclusions, and Discussions: Results of the TUG show an increased completion time for all trials performed with the dEMG powered prosthesis as compared to the passive prosthesis (Fig. 1). However, the dEMG TUG completion time decreased by 31% between the initial TUG and final TUG (Fig. 1). The decrease in completion time implies rapid learning and may produce similar results to the passive prosthesis with training. The dEMG Stroop test showed a decrease in percent accuracy and an increase in average reaction time for both seated and walking conditions (Fig. 1), indicating a greater cognitive load as compared to a passive prosthesis. The stand and reach showed dEMG increased forward reach by nearly 30% and residual (left) side reach by 16% compared to use of a passive prosthesis (Fig. 2), suggesting dEMG may provide a substantial benefit in similar activities of daily life. The expected push perturbation task revealed dEMG created more symmetry between limbs in comparison to a passive prosthesis (Fig. 3). Additionally, all sway stances showed an increase in limb symmetry with the dEMG prosthesis compared to a passive prosthesis (Fig. 4). An increase in limb symmetry may indicate a decrease in reliance on compensatory strategies from the intact limb. As the participant only had a few trials at each task, the decrease in TUG times and more identical limb symmetry signifies training could have a factor on device performance. Future research should focus on evaluating the effect of formal training on clinical and biomechanical task performance related to balance and postural stability, as well as cognitive load.
This study expands on previous work as a preliminary evaluation of additional clinical tasks without any training with the powered device. We found that using the dEMG prosthesis improved forward and residual side reach, but resulted in slower TUG times as compared to passive. However, the significant difference between initial and final TUG times with the dEMG prosthesis demonstrates the user's performance improvement with a few walking trials. Implementing a training protocol for the dEMG powered prosthesis could enhance performance in all tasks, indicating potential improvements in postural stability, cognitive load, and mobility.

Acknowledgements (Optional):