Abstract: Objective　To explore the application effect of artificial intelligence (AI) automated landmark recognition technology in assisting the teaching of orthodontic cephalometric analysis.Methods　Twenty 5-year program undergraduate students majoring in stomatology from Shanghai Jiao Tong University School of Medicine were selected and divided into traditional and AI-assisted groups, with 10 students in each group. The Uceph digital cephalometric software was used to perform the orthodontic cephalometry in both groups. Manual landmark detection was used in the traditional group and AI automated landmark detection technology was added after manual detection in the AI-assisted group. After training, each student independently completed 10 cephalometric analyses and relevant questionnaires.Results　Compared with the teacher's reference value, the measurement differences of SNA and SN-MP were significantly smaller in the AI-assisted approach than those in the traditional group, the difference was statistically significant (P＜0.05). The questionnaire results showed that the score of the AI-assisted automated landmark detection (8.48±0.59) points was higher than that of the traditional group (6.26±1.21) points, which was statistically significant (P＜0.001). The advantages of the AI automated landmark detection method as evaluated by the students were mainly "improving classroom interest", "promoting the communication and interaction between teachers and students", and "understanding the craniomaxillofacial structure more clearly and accurately".Conclusion　The technique of AI-assisted automated landmark detection has achieved good teaching effects in cephalometric practice teaching and student acceptance, which is worthy of further improvement and promotion.

Key words: cephalometric analysis, artificial intelligence, automated landmark detection, orthodontics, practice teaching, teaching evaluation