TITLE:
Validity of Artificial Intelligence Models in Orthodontic Diagnosis: A Systematic Review
AUTHORS:
Kenza Khamlich, Farid Bourzgui
KEYWORDS:
Artificial Intelligence (AI), Orthodontics, Orthodontic Diagnosis, Neural Networks, Machine Learning
JOURNAL NAME:
Open Access Library Journal,
Vol.13 No.4,
April
23,
2026
ABSTRACT: Objective: This systematic review aimed to evaluate the validity of artificial intelligence (AI)-based models applied to orthodontic diagnosis. Materials and Methods: A comprehensive electronic search was conducted in five databases—PubMed, ScienceDirect, Google Scholar, Web of Science, and the Cochrane Library—using the MeSH terms artificial intelligence, orthodontics, orthodontic diagnosis, neural networks, and machine learning. After applying predefined inclusion and exclusion criteria, nine studies were selected for full-text review and critical appraisal. Results: The initial search identified 325 studies related to AI in orthodontic diagnosis, and after screening titles and abstracts, 52 full-text articles were assessed for eligibility, of which eleven met the inclusion criteria. The included studies evaluated various AI algorithms for their diagnostic accuracy and clinical applicability. Conclusion: The evidence suggests that AI can enhance diagnostic accuracy and efficiency in orthodontics, offering significant potential to improve diagnosis, decision-making, treatment monitoring, and prediction of treatment outcomes. However, further research with standardized methodologies and larger clinical datasets is needed to validate the reliability and generalizability of AI-based diagnostic models in orthodontic practice.