An mHealth-Based Learning Intervention for Junior Nurses: Effects on Cognitive Outcomes and Self-Efficacy under Multimodal Generative AI in Medical Imaging Procedures

The integration of mobile health (mHealth) technologies and artificial intelligence (AI) is transforming clinical education and healthcare training environments. However, many existing digital learning systems rely on static instructional materials and lack adaptive mechanisms capable of responding to learners’ behavioral patterns. To address this limitation, this study developed a multimodal generative AI–enhanced mHealth learning system designed to support junior nurses during medical imaging preparation procedures. A quasi-experimental pretest–posttest study was conducted with 60 newly recruited nurses at a tertiary medical center in Taiwan. Participants were assigned to a control group receiving conventional training or an experimental group using the AI-assisted mHealth platform for four weeks. The system integrates generative instructional content and adaptive learning analytics, using multimodal inputs—including text, visual demonstrations, and user interaction logs—to generate personalized feedback during learning tasks. Results showed that the experimental group achieved significantly higher cognitive scores (p < 0.001) and stronger knowledge retention compared with the control group. Improvements were also observed in learning attitude, self-efficacy, and imaging task accuracy, with task completion increasing from 88.7% to 100%. These findings demonstrate that multimodal generative AI–enabled mHealth systems can enhance procedural learning, learner confidence, and operational accuracy in clinical training environments, highlighting the potential of AI-assisted learning platforms to support intelligent healthcare education.