Propulsion Modulation Methods in People Post-Stroke during Resistive Ankle Exosuit Use

• Published in: 2024 10th IEEE RAS/EMBS International Conference for Biomedical Robotics and Biomechatronics (BioRob) pp. 15 - 22
• Main Authors: Swaminathan, Krithika, Lee, Christina, Schmitz, Dylan G., Baker, Teresa, Chin, Andrew, Wendel, Nicholas, Awad, Louis N., Ellis, Terry D., Thelen, Darryl G., Walsh, Conor J.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.09.2024
• Subjects: Ankle; Biomechanics; Hip; Modulation; Muscles; Propulsion; Robot kinematics; Stroke (medical condition); Training
• ISSN: 2155-1782
• DOI: 10.1109/BioRob60516.2024.10719898

Locomotion requires careful coordination across the various joints and muscles of the body, which can be disrupted after neuromotor injuries such as stroke. People post-stroke often have weakness in their paretic, or more impaired, ankle plantarflexors and a corresponding reliance on the hip joint to generate sufficient forward propulsion. The field of robotic rehabilitation has developed wearable systems that provide joint-and task-specific training for survivors of stroke, and in turn, increase use of the ankle muscles. However, capturing ankle use at the plantarflexor level remains a challenge with conventional tools given the unknown relative contributions of the dorsiflexor muscles. Moreover, variability across individuals complicates the interpretation of user response to these robotic interventions. In this work, we used standard biomechanical analysis as well as shear wave tensiometry in five people post-stroke to gain insight into user-specific ankle and hip adaptations in response to three levels of targeted plantarflexion exosuit resistance. We show that at a group and individual-level, evidence suggests a shift in biomechanical strategy from relying on the hip to using the ankle to modulate propulsion, with a subset of participants completely shifting to the ankle by the end of training. This work represents a step towards exploring more individualized methods for characterizing user response during adaptation to wearable robotic training interventions.