The Impact of an AI-Enhanced Teaching Model on Cultivating Clinical Reasoning in Ultrasound Diagnosis among Medical Students.

OBJECTIVE: To explore the application effect of a new teaching model empowered by artificial intelligence (AI) in the cultivation of clinical thinking in ultrasound diagnosis. METHODS: A total of 67 medical imaging and imaging technology interns from January 2023 to June 2025 were randomly divided into an experimental group (n = 34) and a control group (n = 33). Students in the experimental group generated ultrasound reports based on real cases through an ultrasound workstation and a clinical work system. AI technology was used to analyze students' diagnostic thinking, description methods, etc., and automatically generate differential diagnosis suggestions. Summaries and feedback were provided after each stage of learning. The control group adopted the traditional lecture + image analysis teaching mode. The differences in theoretical scores, the "Ultrasound Clinical Thinking Ability Assessment Scale", and the "Ultrasound Clinical Thinking Intelligent Teaching Satisfaction Questionnaire" between the two groups were ultimately evaluated. RESULTS: Compared with the control group, the AI-empowered teaching model could enhance students' clinical thinking abilities such as information integration, differential diagnosis, and diagnostic decision-making (P < 0.05). Students taught through the AI-empowered teaching model had higher teaching satisfaction in "professional ability" and "thinking improvement" than the control group (P < 0.05). However, there were no significant differences between the two groups in theoretical scores, differential diagnosis thinking ability, and teaching mode satisfaction (P > 0.05). CONCLUSION: The AI-enhanced novel teaching model was associated with improved clinical reasoning skills of ultrasound medicine trainees and increase their teaching satisfaction, providing a novel approach for ultrasound medical education. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at https://doi.org/10.1007/s40670-026-02781-w.