Effects of 8-Week Wearable Ultrasound Visual Feedback Training on Walking Performance and Joint Coordination of Older Adults with Chronic Stroke: A Randomized Controlled Trial

1. Introduction

Stroke frequently causes paretic ankle dorsiflexor weakness, impairing gait coordination and increasing fall risks. While a single session of wearable ultrasound visual feedback (UVF) training has been shown to improve isometric ankle dorsiflexion strength in chronic stroke survivors, the transfer of these effects to overground walking is unknown. This study investigated effects of an 8-week UVF training program on walking performance and inter-joint coordination in community-dwelling older adults with stroke. 

2. Method 

Thirty-four older adults with chronic stroke were randomized (17 control, 17 experimental). Both groups performed seated isometric tibialis anterior training three times weekly for eight weeks; the experimental group received additional real-time UVF of TA muscle contraction patterns, provided by a wearable wireless ultrasound device. Outcomes were assessed during self-selected overground walking at baseline, mid- (12 sessions), and post-intervention (24 sessions). Walking performance was quantified via spatiotemporal parameters. Knee-ankle coordination in sagittal, coronal, and transverse planes was classified using modified vector coding across seven gait sub-phases. Two-way mixed ANOVAs analysed training effects.

3. Results

Significant main effects of time were found for gait speed, stride length, and step length (p<0.05), indicating improved walking performance in both groups. Significant Time×Group interactions occurred during sagittal early swing and transverse mid-swing (p≤0.025). Post-hoc analyses revealed that for sagittal early swing, the experimental group significantly reduced ankle-dorsi/plantarflexion-dominated patterns over time (pre- vs. post- MD=-9.9%, p=0.027), resulting in a significantly lower frequency compared to controls at post-intervention (MD=-15.5%, p=0.002). In contrast, the interaction for transverse mid-swing was driven by a lower baseline frequency of ankle-rotation-dominated patterns in the experimental group (MD=-15.6%, p=0.015), with no significant longitudinal changes observed.

4. Conclusions

While general walking performance improved in both groups, the addition UVF training specifically modulated inter-joint coordination by reducing ankle-dominated patterns during the swing phase, suggesting its potential for targeted gait rehabilitation.

Acknowledgement: This work was supported by the Hong Kong Research Grants Council (RGC) - Early Career Scheme (ECS) (Reference number: 25100523).