Managing a frightened child in a dental chair is one of the hardest skills for a student dentist to learn. It demands fast thinking, careful language, and the ability to stay calm while a young patient resists treatment. Teaching that skill at scale has always been difficult, but researchers at the National University of Singapore think they have found a way to make it more efficient.
According to Phys.org, a study published in the scientific journal JMIR Medical Education found that dental students' behavior guidance skills can be accurately assessed using only written transcripts of their clinical sessions. Educators have traditionally relied on video recordings of live sessions to evaluate student performance, a process that is time-consuming and requires significant faculty hours.
The research was conducted by Dr. Ishreen Kaur, Dr. Gabriel Lee, and Associate Professor Hu Shijia from the NUS Faculty of Dentistry. Their findings showed that evaluations based solely on text transcripts produced results that were comparable to assessments drawn from full video-recorded sessions. If the written record of what was said is enough to judge how a student handled a difficult interaction, the hours spent reviewing footage could be redirected elsewhere.
The stakes in pediatric dentistry are high. Previous studies have found that dental students experience stress levels three times higher than those of seasoned specialists when handling young or uncooperative patients. Developing the confidence to communicate clearly in those moments typically requires direct, personalized feedback from faculty during or after live clinical sessions. The problem is that faculty resources are often stretched thin, especially in high-enrollment programs.
Hu described what the findings could mean for training. "These findings allow us to enhance training for our dental students by providing more personalized and insightful feedback on their interactions with young patients. With a better understanding of each student's needs, we can better tailor our training to help them grow into more confident and effective dentists, ultimately providing better care for young patients," Hu said.
The shift to transcripts also opens a door that video does not. Because text can be processed digitally, it becomes possible to use artificial intelligence tools, specifically large language models, to analyze student interactions and generate feedback automatically. The NUS team sees this as a long-term path toward what they described as a virtual mentor system, one that could support students in building communication skills in pediatric care while freeing faculty to focus on higher-level guidance and more complex cases.
The researchers noted that the approach is not limited to dentistry. They suggested it could be adapted to other health care fields, including nursing and medicine, wherever training involves managing child patients and requires strong communication skills under pressure.
The team plans to carry out a follow-up study to assess how effective AI-generated feedback is when applied to dental student transcripts. That study will help determine whether the virtual mentor concept is practical or still years away from clinical use.
