Currently submitted to: JMIR Serious Games
Date Submitted: Aug 9, 2026
Open Peer Review Period: Aug 11, 2026 - Oct 6, 2026
(currently open for review)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Generative AI-Assisted Mixed Reality for Adaptive Cognitive-Motor Rehabilitation in Chronic Stroke: A Randomized Controlled Trial
ABSTRACT
Background:
Chronic stroke often results in cognitive-motor interference (CMI) and pathological ankle co-contraction that limit overground gait independence.
Objective:
This study evaluated a generative AI-assisted mixed reality (MR) dual-task rehabilitation program designed to reduce pathological ankle co-contraction and support gait automaticity in chronic stroke survivors.
Methods:
Fifty participants were randomly allocated to a Generative AI-Assisted Mixed-Reality Zone (GAZE) group or a Traditional Rehabilitation Therapy (TRT) group (n=25 each) for a 4-week intervention comprising 12 sessions (360 minutes total). In the GAZE group, a large language model dynamically adapted visual-cognitive task presented through an MR head-mounted display during overground walking within a human-verified adaptive staircase framework. Primary outcomes were tibialis anterior and gastrocnemius activation normalized to reference voluntary contraction and the ankle co-contraction index (CI), assessed using surface electromyography. Secondary outcomes included functional mobility (6MWT), activities of daily living (MBI), dual-task cost, and HMD-derived upper-limb kinematics.
Results:
Significant group × time interactions favoring the GAZE group were observed for tibialis anterior activation, gastrocnemius activation, and CI. Both groups demonstrated comparable improvements in general functional outcomes, including the 6MWT, MBI, falls efficacy, and quality of life. Within the GAZE group, dual-task cost and execution speed also improved significantly following the intervention.
Conclusions:
These preliminary findings suggest that generative AI-assisted MR dual-task training may provide additional benefits for distal neuromuscular control beyond conventional rehabilitation. Clinical Trial: Clinical Research Information Service (CRIS), Republic of Korea (registration no. KCT0011952; registered on 10 May 2026)
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.