Enhancing dental clinical education through AI: insights into post-clinical summaries

Abstract

BACKGROUND: Artificial Intelligence (AI) is increasingly being introduced into clinical education, including dentistry, as a supplement to traditional instructor-led reflection and feedback. Post-clinical summaries-structured end-of-day reviews of patient cases, decision-making, safety concerns, and areas for improvement-are a critical part of this training process. OBJECTIVE: This study explored how trainees perceive the usefulness of AI tools such as ChatGPT and DEEP SEEK in supporting post-clinical summaries, with respect to knowledge consolidation, diagnostic reasoning, teacher-student communication, and professionalism. METHODS: We conducted a cross-sectional perception survey of 54 participants (undergraduate interns through resident physicians) who had experience with post-clinical summaries. The structured questionnaire used Likert-type items to assess perceived value of AI in four domains. Descriptive statistics were used to summarize responses. RESULTS: Most respondents reported that AI tools help them review theoretical knowledge after clinical work (85.19% agreed/strongly agreed) and clarify diagnostic reasoning (74.08%). Many perceived AI as improving the depth and personalization of teacher-student communication (74.07%) and enhancing confidence in asking questions (72.22%). By contrast, responses regarding non-technical competencies-such as ethical awareness, responsibility, and professional judgment-were more mixed, with many respondents selecting neutral options. Overall, 77.78% of respondents agreed that AI is a valuable resource for improving post-clinical summary activities, and 68.52% would recommend integrating AI into clinical education. CONCLUSIONS: Participants generally perceived AI as a helpful adjunct for reinforcing clinical knowledge, supporting diagnostic reasoning, and facilitating communication. Perceived benefits for professionalism, ethics, and responsibility were less clear. Because this study used self-reported perceptions from a single setting, without qualitative data or inferential statistics, the findings should be interpreted as exploratory. Future work should include qualitative interviews, objective performance measures, and longitudinal follow-up to determine whether AI-supported post-clinical summaries translate into measurable educational outcomes.

Ähnliche Arbeiten