<p>The preventive medicine education mainly relies on traditional teaching strategies, survey-based evaluations, or conventional AI models that lack adaptive learning support and multimodal data integration. These approaches often fail to provide personalized curriculum adjustments based on dynamic student behavior and physiological feedback. To address these limitations, this research aims to optimize the preventive medicine teaching curriculum by developing a thoughtful, data-driven system that enhances learning effectiveness, student engagement, and health awareness. The proposed system integrates IoT-enabled wearable devices, smart classroom sensors, and digital health platforms to collect physiological, behavioral, and academic data from 2000 medical student records in a smart preventive medicine education dataset. These datasets are preprocessed using a median filter to remove sensor noise and min–max normalization to standardize feature ranges, followed by feature extraction using Principal Component Analysis (PCA) to minimize dimensionality and preserve the most significant learning patterns. An advanced AI technique, including machine learning (ML) algorithms such as Intelligent Migrating Birds Optimized Extreme Gradient Boosting (IMB-XGBoost) models, analyzes learning behaviors, predicts performance trends, and assesses emotional adaptability. The system dynamically personalizes curriculum and teaching based on predictions. Python 3.10 experiments show the AI–IoT curriculum improves emotional regulation, self-directed learning, and comprehension, achieving 94.3% of accuracy and attaining superior values in other metrics. The optimized system not only fosters academic achievement but also strengthens preventive health literacy and critical thinking abilities among students. This research highlights the potential of intelligent technologies to enable scalable, adaptive, and personalized preventive medicine education.</p>