Application of AI-based virtual standardized patients in physician-patient communication training: a study based on the SEGUE framework

Background: Developing effective doctor-patient communication skills is a critical component of medical education. The SEGUE framework offers a structured and systematic approach for teaching and assessing communication competence. However, traditional standardized patient (SP) training is resource-intensive, time-consuming, and difficult to scale for large student cohorts. AI-based virtual standardized patients (AI-VSPs) have emerged as a promising alternative, providing repeatable, accessible, and scalable training opportunities. This study aimed to evaluate the effectiveness and feasibility of AI-VSPs within SEGUE-based communication training for medical students, compared with traditional SPs. Methods: = 42) for 2 h following a 3-week medical ethics course. Communication performance was assessed using SEGUE total scores, Content and Process subscales, training frequency was recorded, and participant feedback was analyzed qualitatively. Results: = 0.03). Qualitative feedback highlighted AI-VSPs' repeatability, convenience, and realistic interactions, with fewer negative comments than traditional SPs. Conclusion: SEGUE-based simulation training effectively enhances medical students' communication skills. AI-VSPs offer scalable, repeatable, and practical advantages, helping learners complete communication tasks thoroughly and providing targeted feedback, but they do not necessarily improve real-world communication competence such as empathy or relational skills.