AI-Based Exergaming as a Digital Assessment Tool for Post-Stroke Upper-Limb Function: Protocol for a Feasibility...

Area of Science:

• Neurorehabilitation
• Digital Health
• Biomedical Engineering

Background:

• Accurate upper limb motor function assessment is crucial for stroke survivor rehabilitation.
• The Fugl-Meyer Assessment for Upper Extremity (FMA-UE) is standard but time-consuming and variable.
• Computer vision exergames offer a scalable, objective, and remote assessment alternative.

Purpose of the Study:

• To evaluate the feasibility and preliminary validity of a computer vision-based exergame for assessing post-stroke upper limb motor function.
• To investigate correlations between kinematic gameplay features and FMA-UE scores.
• To establish the potential of this digital tool for remote and objective motor assessment.

Main Methods:

• A cross-sectional pilot study involving 15-30 individuals with chronic stroke.
• Participants engaged in a markerless camera-based exergame using MediaPipe hand tracking.
• Kinematic features (range of motion, hand angle, jitter index) were extracted and correlated with FMA-UE scores.

Main Results:

• Expected moderate to strong correlations between gameplay metrics (range of motion, jitter index) and FMA-UE scores.
• Anticipated high usability scores (System Usability Scale) indicating feasibility.
• Preliminary data suggests potential for objective and scalable motor assessment.

Conclusions:

• The computer vision exergame shows potential as an accessible tool for objective stroke rehabilitation assessment.
• Validation could lead to reduced clinical workload and enable remote patient evaluations.
• Future research will focus on test-retest reliability and home-based implementation.