Validation and Generalisability of The Use of Artificial Intelligence(AI) in Orthodontic Diagnosis: A Scoping Review

This scoping review aims to produce a comprehensive descriptive mapping on the validation of AI models used for orthodontic diagnosis. Artificial intelligence (AI) is used increasingly in orthodontics for tasks such as detecting cephalometric landmarks, automated segmentation, classifying malocclusions, and assessing growth. Although many studies report promising results, it is unclear how these AI models have been evaluated or whether their performance is consistent across various datasets, populations, and clinical situations (generalisability). This review will focus on highlighting gaps in the literature to provide insights for future research and help guide the safe, ethical and effective use of AI in everyday orthodontic practice.