AI-Based Virtual Standardized Patients Improve Oral Mucosal Disease Training in Fourth-Year Dental Students

Purpose: History-taking is essential for the diagnosis of oral mucosal diseases, and subsequent medical record writing is a fundamental clinical skill for dental students. While standardized patients (SPs) have traditionally been used to improve these abilities, their application requires considerable human, material, and time resources. This study aimed to evaluate an AI-based virtual SP system—integrated with lesion images and examination results—as a convenient and feasible alternative for teaching oral mucosal disease history-taking and medical record writing. Methods: Fourth-year undergraduate dental students attended an oral mucosal disease course and, after each lecture, completed simulated interviews with an AI-based virtual SP using voice interaction. Lesion photographs and auxiliary examination results were provided to aid diagnosis. Students wrote medical records online within 20 min based on the collected history. A comprehensive assessment was conducted at the end of the course. Performance in history-taking and medical record writing during the first AI-based session (T1) was compared with the final assessment (T2) using paired analyses. Student satisfaction with the system was assessed via an online questionnaire. Results: Paired comparisons showed a significant mean improvement in history-taking scores (+18.64 points; 95% CI: 15.42–21.87). Overall medical record writing scores also increased, particularly in completeness and format standardization. Most students agreed that integrating the AI-based virtual SP into the curriculum improved their oral mucosal disease history-taking and medical record writing skills, and they supported its application in dental education. Conclusions: Incorporating AI-based virtual SPs into oral mucosal disease teaching can effectively improve key clinical documentation skills while offering a time- and resource-efficient solution for dental education. Trial Registration: Approved by the Fourth Military Medical University IRB protocol number #2024YB31.