AI-Assisted Case-Based Learning and Flipped Classroom to Improve Clinical Decision-Making: A Randomized Controlled Trial in Reproductive Medicine
Description
In this study, we hypothesized that an instructional model integrating artificial intelligence assisted case based learning with a flipped classroom would outperform traditional lecture based instruction in enhancing clinical knowledge, decision making skills, and learning engagement among reproductive medicine residents. To test this, fifty eligible obstetrics and gynecology residents were randomly assigned to either the AI assisted CBL and flipped classroom group or a traditional control group. Data were collected through standardized theoretical knowledge examinations, Mini Clinical Evaluation Exercise assessments, and Objective Structured Clinical Examinations administered after the intervention, as well as learner feedback surveys using a five‑point Likert scale on motivation, clinical thinking, self directed learning, and perceived course effectiveness. The results demonstrated that participants in the AI assisted group achieved significantly higher post course theoretical test scores compared with controls, indicating superior knowledge acquisition. Objective clinical assessments showed that the AI assisted group also exhibited markedly better performance in medical interviewing, physical examination, and overall clinical competence, reflecting improved practical abilities and decision making in simulated clinical scenarios. Furthermore, subjective evaluations revealed greater improvements in learning motivation, development of clinical thinking, and self directed learning in the AI assisted group, suggesting that this instructional model enhanced learner engagement and intrinsic motivation. These findings support the interpretation that the AI assisted CBL and flipped classroom format facilitates deeper integration of theoretical knowledge with clinical application, promotes active learning, and accelerates the development of essential clinical competencies. Collectively, the data provide evidence for the adoption of technology supported, learner centered teaching strategies in reproductive medicine education and offer a foundation for future instructional design and broader implementation.