Objectives: To investigate the current clinical applications and diagnostic performance of artificial intelligence (AI) in dental and maxillofacial radiology (DMFR). Methods: Studies using applications related to DMFR to develop or implement AI models were sought by searching five electronic databases and four selected core journals in the field of DMFR. The customized assessment criteria based on QUADAS-2 were adapted for quality analysis of the studies included. Results: The initial electronic search yielded 1862 titles, and 50 studies were eventually included. Most studies focused on AI applications for an automated localization of cephalometric landmarks, diagnosis of osteoporosis, classification/segmentation of maxillofacial cysts and/or tumors, and identification of periodontitis/periapical disease. The performance of AI models varies among different algorithms. Conclusion: The AI models proposed in the studies included exhibited wide clinical applications in DMFR. Nevertheless, it is still necessary to further verify the reliability and applicability of the AI models prior to transferring these models into clinical practice.

No abstract

No abstract

Learning bone anatomy of the skull is a complex topic involving three‐dimensional information. The impact of the use of human dry skulls and cone beam computed tomography (CBCT) imaging was investigated in the teaching of undergraduate dental students. Sixty‐four first‐year students in the University of Hong Kong were randomly divided into eight groups. Four teaching methods were tested: (1) CBCT followed by standard lecture, (2) CBCT followed by lecture with skulls, (3) standard lecture followed by CBCT, and (4) lecture with skulls followed by CBCT. After each, students were given a multiple‐choice questionnaire to assess their objective learning outcome (20 questions) and a questionnaire for their subjective satisfaction (10 statements). Surveys were assessed with Cronbach's alpha, Kendall's tau‐b, and principal components analysis. Data were analyzed with Student's t‐test and a one‐way ANOVA (significance α = 0.05). Standard lecture followed by CBCT showed the highest learning outcome score (81.6% ± 14.1%), but no significant difference was present among four teaching methods. Cone beam computed tomography followed by lecture with skulls scored the highest overall subjective satisfaction (4.9 ± 0.8 out of 6), but no significant difference was present among teaching methods. Nevertheless, students' perception of learning was positively influenced by the use of skulls (P = 0.018). The timing of administration of the CBCT did not affect students' subjective satisfaction or objective learning outcome. Students perceived to learn more by using skulls, but their objective learning outcomes were not significantly affected. A discrepancy seems to exist between students' perception of learning and their effective performance.