2型糖尿病高リスク者に対する自宅介入のための単眼カメラベースのモバイルエクササイズゲームの開発と評価：ランダム化比較試験

Exergames have emerged as effective interventions for promoting physical activity and preventing type 2 diabetes (T2D). Kinect-based exergames have demonstrated improvements in exercise adherence and health outcomes, but their high cost and reliance on specialized hardware hinder widespread home-based adoption. Recent advances in computer vision now enable monocular camera-based systems, offering a potentially cost-effective and scalable alternative for promoting physical activity at home.

This study aimed to evaluate the feasibility and user experience of monocular camera-based exergames as a home-based intervention for individuals at risk for T2D.

Forty-five community-dwelling individuals at high risk for T2D (mean age 47.12, SD 6.92 years) were recruited and randomized into three groups (n=15 each): (1) control group (traditional offline exercise), (2) Kinect group (Kinect-based exergame), and (3) monocular group (monocular camera-based exergame). Participants engaged in a 10-minute intervention once per week for 7 weeks. Data were collected at 3 time points: baseline (exercise performance: heart rate and perceived fatigue), postintervention (exercise performance and user experience, including game experience and intrinsic motivation), and follow-up (user engagement and qualitative feedback). One-way ANOVA was used for data analysis.

Exercise performance was comparable across all groups, with no significant differences in heart rate (P=.76) or fatigue levels (P=.25). However, participants in the monocular group reported significantly lower fatigue than those in the control group (P=.04). Intrinsic motivation was significantly higher in both the Kinect (mean 35.13, SD 3.20) and monocular (mean 34.00, SD 4.41) groups than in the control group (mean 26.06, SD 1.87; P<.001), with no significant difference between the 2 exergame groups (P=.44). While most user experience measures showed no significant differences, the monocular group reported a higher perceived challenge (mean 3.45, SD 0.51) than the Kinect group (mean 2.96, SD 0.39; P=.09). Additionally, the monocular group exhibited higher engagement, as evidenced by more frequent use, fewer challenges, and a greater intention to continue using the system.

Monocular camera-based exergame is a feasible and effective solution for promoting physical activity in individuals at risk for T2D. It offers motivational and experiential benefits similar to Kinect-based systems but requires less costly and more accessible equipment. These findings suggest that monocular systems have strong potential as scalable tools for home-based chronic disease prevention.