中国口腔颌面外科杂志

中国口腔颌面外科杂志 ›› 2026, Vol. 24 ›› Issue (4): 406-412.doi: 10.19438/j.cjoms.2026.04.014

• 医学教育 • 上一篇    下一篇

AI分割下颌第三磨牙在口腔住院医师规范化培训中的应用与评价

周琴1, 李柯萱1, 孙韫2, 房笑1, 王欣倍3, 姬常凯4, 杜观环5   

  1. 1.上海交通大学医学院附属第九人民医院 口腔外科,上海交通大学口腔医学院,国家口腔医学中心, 口腔疾病国家临床医学研究中心,上海市口腔医学重点实验室,上海市口腔医学研究所,上海 200011;
    2.上海交通大学口腔医学院,上海 200011;
    3.上海交通大学医学院附属第九人民医院 规范化培训办公室,上海 200011;
    4.上海交通大学自动化与感知学院,上海 200240, 5.上海交通大学医学院附属第九人民医院 口腔黏膜病科,上海交通大学口腔医学院,国家口腔医学中心, 国家口腔疾病临床医学研究中心,上海市口腔医学重点实验室,上海市口腔医学研究所,上海 200011
  • 收稿日期:2026-01-28 修回日期:2026-04-08 发布日期:2026-08-05
  • 通讯作者: 杜观环,E-mail: dgh-09@163.com
  • 作者简介:周琴（1982—）,女,博士,副主任医师,E-mail: qin_zq@163.com
  • 基金资助:
    上海交通大学医学院附属第九人民医院教学项目（JYJX03202413,JYJX01202501）; 上海交通大学医学院附属第九人民医院临床研究助推计划（JYLJ202417）; 高校兼职辅导员开展口腔医学生职业生涯规划教育的路径研究（kqjzfdy002）

Investigation on learning application and cognitive evaluation of AI segmentation of panoramic images of mandibular third molars among standardized training residents in stomatology

Zhou Qin1, Li Kexuan1, Sun Yun2, Fang Xiao1, Wang Xinbei3, Ji Changkai4, Du Guanhuan5   

  1. 1. Department of Oral Surgery, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine;College of Stomatology, Shanghai Jiao Tong University;National Center for Stomatology;National Clinical Research Center for Oral Diseases;Shanghai Key Laboratory of Stomatology;Shanghai Research Institute of Stomatology. Shanghai 200011;
    2. College of Stomatology, Shanghai Jiao Tong University. Shanghai 200011;
    3. Standardized Training Office, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine. Shanghai 200011;
    4. School of Automation and Intelligent Sensing, Shanghai Jiao Tong University. Shanghai 200240;
    5. Department of Oral Mucosal Diseases, Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine;College of Stomatology, Shanghai Jiao Tong University;National Center for Stomatology;National Clinical Research Center for Oral Diseases;Shanghai Key Laboratory of Stomatology;Shanghai Research Institute of Stomatology. Shanghai 200011, China
  • Received:2026-01-28 Revised:2026-04-08 Published:2026-08-05

PDF (PC)

30

摘要： 目的：分析口腔住院医师规范化培训（口腔规培）学生对人工智能（artificial intelligence,AI）分割下颌第三磨牙全景图像的学习效果与反馈,为教学中合理应用 AI、提升学习效率与教学质量提供依据。方法：采用问卷星匿名问卷调查上海交通大学医学院附属第九人民医院口腔规培学生,内容包括基本信息、基于病例的影像判定（下颌第三磨牙原始全景片与AI分割图像各4组）,以及对AI的认知、应用需求与教学期待等。将学生对病例影像的判读结果与专家组标准进行比较。结果：共收到65名学生的有效问卷。在牙根及下牙槽神经管的关系判读方面,AI分割组均较全景片组与专家组标准一致性更高（P<0.05）,但拔牙难度自评分与专家标准一致性方面无统计学差异（P>0.05）。95.38%的学生愿意在未来更多接触AI,95%以上学生支持AI在口腔教育中的应用与发展。结论：AI分割图像可显著优化下颌第三磨牙影像学习的解剖可视化效果与学习体验,学生接受度高。建议AI在教师指导下与传统教学融合,构建“AI+教师+学生”互动教学模式,并向个性化学习、虚拟仿真教学拓展。

关键词: 人工智能, 下颌第三磨牙, 口腔教学, 认知调查

Abstract: PURPOSE: To analyze the learning effect and feedback of standardized training residents in stomatology on artificial intelligence (AI) segmentation of panoramic images of mandibular third molars, so as to provide evidence for the rational application of AI in teaching and the improvement of learning efficiency and teaching quality. METHODS: An anonymous online questionnaire using Wenjuanxing was distributed to stomatology (dental) residents receiving standardized training at Shanghai Ninth People's Hospital, Shanghai Jiao Tong University School of Medicine. The questionnaire included general information, case-based image evaluation (4 sets of original panoramic radiographs and 4 sets of AI-segmented images of mandibular third molars), as well as participants' cognition of AI, application demands and teaching expectations. The residents' image interpretation results were compared with the expert panel criteria. RESULTS: A total of 65 valid questionnaires were collected. In the interpretation of the positional relationship between tooth roots and the inferior alveolar nerve canal, the AI segmentation group showed significantly higher consistency with the expert criteria than the panoramic radiograph group (P<0.05). No significant difference was found in the consistency of self-rated extraction difficulty with expert standards between the two groups (P>0.05). A total of 95.38% of the residents were willing to have more access to AI in the future, and over 95% of the residents supported the application and development of AI in dental education. CONCLUSIONS: AI segmentation of images can significantly enhance the anatomical visualization and learning experience for the study of third mandibular molars, with high acceptance among learners. It is recommended that AI be integrated with traditional teaching under the guidance of instructors to establish an interactive teaching model of "AI + instructor + student", and to expand into personalized learning and virtual simulation teaching.

Key words: Artificial intelligence, Mandibular third molar, Dental teaching, Cognitive survey

中图分类号: