AI Empowers Traditional Physical Education Teaching in Higher Vocational Education: Exploring New Paths for the Integration of Health Education

This study evaluates an AI-guided atomization sensing system designed to integrate adaptive health education into vocational physical education. A quasi-experimental design was conducted with 356 participants allocated to either an intervention group (n = 178) or a control group (n = 178). The system used wearable atomization sensors (monitoring respiratory patterns, ambient temperature, and motion) connected to an AI engine that applied rule-based and machine-learning logic to adjust instructional content in real time-for example, reducing complexity or switching to audiovisual demonstrations when signs of fatigue were detected. Sessions were delivered in classroom settings twice per week for six weeks, with surveys administered immediately before and after the intervention. Five outcomes were assessed using validated instruments: knowledge retention, engagement, behavioral intention, self-efficacy, and technology acceptance. Statistical analyses included ANOVA, paired-sample t-tests, Pearson correlations, and χ² tests, with results reported as F(df1, df2), p, η², and 95% confidence intervals where applicable. The intervention group achieved significantly greater improvements than the control group in knowledge retention (Δ = +21.7 vs. +10.6; F(1, 354) = 46.21, p < 0.001, η² = 0.21), engagement (F(1, 354) = 39.87, p < 0.001, η² = 0.18), behavioral intention (F(1, 354) = 42.55, p < 0.001, η² = 0.19), self-efficacy (F(1, 354) = 27.63, p < 0.001, η² = 0.13), and technology acceptance (F(1, 354) = 35.12, p < 0.001, η² = 0.17). These findings demonstrate that combining real-time sensing with AI-guided decision support provides a reproducible, adaptive framework for enhancing health education outcomes in vocational PE settings.