Enhancing Paretic Propulsion Post-Stroke via a Wearable System for Real-Time Unilateral Haptic Feedback of Anterior Ground Reaction Forces

TL;DR

This study demonstrates that wearable inertial measurement units combined with haptic feedback can improve paretic propulsion during overground walking in stroke patients, showing potential for practical gait rehabilitation outside laboratory settings.

Contribution

It introduces a wearable system for real-time AGRF biofeedback during overground walking, extending previous treadmill-based approaches to real-world applications.

Findings

Peak AGRF increased post-training and at retention.

Speed showed significant improvement from baseline.

63% of participants responded positively to the intervention.

Abstract

Gait rehabilitation interventions targeting paretic propulsion can improve walking speed and function in individuals post-stroke. Previous work has demonstrated that real-time biofeedback targeting anterior ground reaction forces (AGRFs) can increase propulsion in individuals post-stroke, however this work was confined to lab-based treadmills, limiting practical utility. Here we investigate the short-term effects of real-time AGRF gait biofeedback during overground walking using wearable inertial measurement units (IMUs) and a haptic feedback device. Eight individuals with chronic post-stroke hemiparesis completed four 3-minute training bouts. During training, faded haptic biofeedback was provided to increase paretic AGRF during terminal stance. Gait biomechanics were assessed before, during, and after training, and during a retention test conducted without biofeedback after a rest period. The primary dependent variable was peak paretic AGRF, while secondary variables included paretic peak trailing limb angle (TLA), step length and walking speed. Compared to baseline, peak AGRF increased post-feedback and at the retention tests. Similar trends were observed in TLA, and step length, although these increases were not statistically significant while speed showed a significant change from baseline. Examining individual participants 63% participants (responders) increased AGRF at retention, while 37% experienced decreases (non-responders). Non-responders had lower physical capability, evidenced by two-minute walk distance at screening and AFO use during training, suggesting this intervention may suit patients with more residual ankle mobility and strength. Nonetheless our results suggest AGRF biofeedback can be implemented in practical settings with wearable systems and is a promising gait training strategy to target propulsive deficits in individuals post stroke.