Evaluation of the Role of Artificial Intelligence in Treatment Planning and Orthodontics: Narrative Review
Downloads
AI is incorporated in almost all stages of the digital orthodontic workflow, from the automated analysis of diagnostic records, the simulation of treatment outcomes, and remote monitoring of tooth movement. In a very short duration, there has been an explosion of models that are published, which has outpaced critical appraisal of their clinical reliability. The aim of this study is to assess the current role of AI in orthodontic treatment planning through the synthesis of reported diagnostic and predictive performance across the main clinical domains. Furthermore, we will examine the influence of task complexity and study design on performance, and finally identify the methodological, ethical, and regulatory barriers that continue to limit routine clinical integration. A narrative review was conducted utilising PubMed/MEDLINE, Scopus, and Web of Science available for articles published from January 2009 to June 2026. Priority was given to systematic reviews, scoping reviews, meta-analyses, and primary diagnostic –accuracy or predictive-modelling studies published in English. The primary authors provided performance metrics, which were extracted without pooling and are presented accordingly. Seven domain were identified cephalometric landmark identification, malocclusion classification, extraction decision support, skeletal maturation assessment, orthognathic outcome prediction, clear aligner planning, remote monitoring. The reported performance follows a consistent gradient determined by the structure of the task and not by its clinical relevance. The tooth segmentation task achieved 98% performance, the two-dimensional landmark detection task achieved 98.3% performance, the binary crossbite classification task achieved 98.6% performance, and the extraction decision task achieved an area under the curve of 91.2%. Multiclass tasks fall to 71–91%. Fine-grained spatial tasks are lowest, with successful detection rates of 67.5% within 2 mm for posteroanterior landmarks and a pooled three-dimensional landmarking error of 2.44 mm. performance also declines when models are across multiple centres: cervical vertebral maturation staging accuracy fell from 98% in a single-centre cohort to 71.1–78.3% in a six-centre cohort involving 3,600 images. AI is delivers its most robust and reproducible benefits in reduced analysis time and reduce inter-observer variability rather than in unambiguous diagnostic superiority over experienced clinicians. given limited external validation, dataset homogeneity, inconsistent reporting and unresolved ethical and medico-legal question, AI is best positioned as a decision-support adjunct requiring clinician oversight at every stage. priorities for the field are multicentre externally validated datasets and prospective designs with patient-relevant endpoints, routine use of explainable AI and adherence to established reporting frameworks.
[1] A. Monill-González, L. Rovira-Calatayud, N. G. d’Oliveira, and J. M. Ustrell-Torrent, “Artificial intelligence in orthodontics: Where are we now? A scoping review,” Orthodontics & Craniofacial Research, vol. 24, pp. 6–15, 2021, doi: 10.1111/ocr.12517.
[2] N. Kazimierczak, W. Kazimierczak, Z. Serafin, P. Nowicki, J. Nożewski, and J. Janiszewska-Olszowska, “AI in orthodontics: Revolutionizing diagnostics and treatment planning—A comprehensive review,” Journal of Clinical Medicine, vol. 13, no. 2, Art. no. 344, 2024, doi: 10.3390/jcm13020344.
[3] S. B. Khanagar et al., “Scope and performance of artificial intelligence technology in orthodontic diagnosis, treatment planning, and clinical decision-making: A systematic review,” Journal of Dental Sciences, vol. 16, no. 1, pp. 482–492, 2021, doi: 10.1016/j.jds.2020.05.022.
[4] S. Azizi, S. Hatampoor, and S. Tahamtan, “Applications of artificial intelligence in diagnosis and treatment planning of orthodontics: A narrative review,” The Saudi Dental Journal, vol. 37, no. 7, Art. no. 70, 2025, doi: 10.1007/s44445-025-00077-0.
[5] R. S. Gracea et al., “Artificial intelligence for orthodontic diagnosis and treatment planning: A scoping review,” Journal of Dentistry, vol. 152, Art. no. 105442, 2025, doi: 10.1016/j.jdent.2024.105442.
[6] F. Preda, N. Elshazly, and R. Jacobs, “AI-enhanced orthodontic treatment planning: A scoping review on evidence-based clinical application with commercial software overview,” Journal of Dentistry, Art. no. 106112, 2025, doi: 10.1016/j.jdent.2025.106112.
[7] C. M. Mörch et al., “Artificial intelligence and ethics in dentistry: A scoping review,” Journal of Dental Research, vol. 100, no. 13, pp. 1452–1460, 2021, doi: 10.1177/00220345211013808.
[8] Z. Jiao, Z. Liang, Q. Liao, S. Chen, H. Yang, G. Hong, and H. Gui, “Deep learning for automatic detection of cephalometric landmarks on lateral cephalometric radiographs using the Mask Region-based Convolutional Neural Network: A pilot study,” Oral Surgery, Oral Medicine, Oral Pathology and Oral Radiology, vol. 137, no. 5, pp. 554–562, 2024, doi: 10.1016/j.oooo.2024.02.003.
[9] P. J. Koch, J. E. Cejudo Grano de Oro, M. Büttner, L. T. Arsiwala-Scheppach, J. De Geer, H. Meyer-Lueckel, and F. Schwendicke, “Deep learning for Angle classification based on intraoral photographs: An interpretability perspective,” BMC Oral Health, vol. 26, no. 1, Art. no. 210, 2026, doi: 10.1186/s12903-025-07550-6.
[10] H. Lee, J. M. Cho, S. Ryu, S. Ryu, E. Chang, Y. S. Jung, and J. Y. Kim, “Automatic identification of posteroanterior cephalometric landmarks using a novel deep learning algorithm: A comparative study with human experts,” Scientific Reports, vol. 13, no. 1, Art. no. 15506, 2023, doi: 10.1038/s41598-023-42870-z.
[11] F. Schwendicke et al., “Deep learning for cephalometric landmark detection: Systematic review and meta-analysis,” Clinical Oral Investigations, vol. 25, no. 7, pp. 4299–4309, 2021, doi: 10.1007/s00784-021-03990-w.
[12] M. Serafin et al., “Accuracy of automated 3D cephalometric landmarks by deep learning algorithms: Systematic review and meta-analysis,” La Radiologia Medica, vol. 128, no. 5, pp. 544–555, 2023, doi: 10.1007/s11547-023-01629-2.
[13] B. Noeldeke, S. Vassis, M. Sefidroodi, R. Pauwels, and P. Stoustrup, “Comparison of deep learning models to detect crossbites on 2D intraoral photographs,” Head & Face Medicine, vol. 20, no. 1, Art. no. 45, 2024, doi: 10.1186/s13005-024-00448-8.
[14] Y. Suhail, M. Upadhyay, A. Chhibber, and Kshitiz, “Machine learning for the diagnosis of orthodontic extractions: A computational analysis using ensemble learning,” Bioengineering, vol. 7, no. 2, Art. no. 55, 2020, doi: 10.3390/bioengineering7020055.
[15] B. Köktürk, H. Pamukçu, and Ö. Gözüaçık, “Evaluation of different machine learning algorithms for extraction decision in orthodontic treatment,” Orthodontics & Craniofacial Research, vol. 27, pp. 13–24, 2024, doi: 10.1111/ocr.12811.
[16] M. H. Mohammed, Z. Q. Omer, B. B. Aziz, J. F. Abdulkareem, T. M. A. Mahmood, F. A. Kareem, and D. N. Mohammad, “Convolutional neural network-based deep learning methods for skeletal growth prediction in dental patients,” Journal of Imaging, vol. 10, no. 11, Art. no. 278, 2024, doi: 10.3390/jimaging10110278.
[17] A. A. Abdulqader, F. Jiang, B. S. Almaqrami, F. Cheng, J. Yu, Y. Qiu, and J. Li, “Deep learning approaches for quantitative and qualitative assessment of cervical vertebral maturation staging systems,” PLOS ONE, vol. 20, no. 5, Art. no. e0323776, 2025, doi: 10.1371/journal.pone.0323776.
[18] P. Farzanegan, M. Zarabadi, S. Najary, M. A. Tahmasbi, and M. Behnaz, “The role of artificial intelligence in orthodontics for determining skeletal age based on cervical vertebra maturation degree: A comprehensive review,” Health Science Reports, vol. 8, no. 11, Art. no. e71487, 2025, doi: 10.1002/hsr2.71487.
[19] P. H. J. de Oliveira et al., “Artificial intelligence as a prediction tool for orthognathic surgery assessment,” Orthodontics & Craniofacial Research, vol. 27, no. 5, pp. 785–794, 2024, doi: 10.1111/ocr.12805.
[20] P. G. Knoops et al., “A machine learning framework for automated diagnosis and computer-assisted planning in plastic and reconstructive surgery,” Scientific Reports, vol. 9, no. 1, Art. no. 13597, 2019, doi: 10.1038/s41598-019-49506-1.
[21] A. A. Almarhoumi, “Accuracy of artificial intelligence in predicting facial changes post-orthognathic surgery: A comprehensive scoping review,” Journal of Clinical and Experimental Dentistry, vol. 16, no. 5, Art. no. e624, 2024, doi: 10.4317/jced.61500.
[22] K. Janáková et al., “The role of artificial intelligence in orthognathic surgery: A scoping review,” Dentistry Journal, vol. 14, no. 5, Art. no. 286, 2026, doi: 10.3390/dj14050286.
[23] T. Wang, H. Chen, G. Song, and B. Han, “Traditional and artificial intelligent methods in predicting maxillofacial soft tissue morphology after orthognathic surgery: A narrative review,” International Journal of Dentistry, vol. 2025, no. 1, Art. no. 6268492, 2025, doi: 10.1155/ijod/6268492.
[24] D. C. Ruiz, S. Mureșanu, X. Du, B. M. Elgarba, R. C. Fontenele, and R. Jacobs, “Unveiling the role of artificial intelligence applied to clear aligner therapy: A scoping review,” Journal of Dentistry, vol. 154, Art. no. 105564, 2025, doi: 10.1016/j.jdent.2025.105564.
[25] N. D. Kravitz, B. Kusnoto, E. BeGole, A. Obrez, and B. Agran, “How well does Invisalign work? A prospective clinical study evaluating the efficacy of tooth movement with Invisalign,” American Journal of Orthodontics and Dentofacial Orthopedics, vol. 135, no. 1, pp. 27–35, 2009, doi: 10.1016/j.ajodo.2007.05.018.
[26] W. Alswajy, H. Baeshen, G. Al-Turki, and F. Alsulaimani, “The reliability of ClinCheck® accuracy before and after Invisalign® treatment—A multicenter retrospective study,” Applied Sciences, vol. 13, no. 8, Art. no. 4670, 2023, doi: 10.3390/app13084670.
[27] G. Rossini, S. Parrini, T. Castroflorio, A. Deregibus, and C. L. Debernardi, “Efficacy of clear aligners in controlling orthodontic tooth movement: A systematic review,” The Angle Orthodontist, vol. 85, no. 5, pp. 881–889, 2015, doi: 10.2319/061614-436.1.
[28] L. Galan-Lopez, J. Barcia-Gonzalez, and E. Plasencia, “A systematic review of the accuracy and efficiency of dental movements with Invisalign®,” The Korean Journal of Orthodontics, vol. 49, no. 3, pp. 140–149, 2019, doi: 10.4041/kjod.2019.49.3.140.
[29] A. Papadimitriou, S. Mousoulea, N. Gkantidis, and D. Kloukos, “Clinical effectiveness of Invisalign® orthodontic treatment: A systematic review,” Progress in Orthodontics, vol. 19, no. 1, Art. no. 37, 2018, doi: 10.1186/s40510-018-0235-z.
[30] F. F. Dai, T. M. Xu, and G. Shu, “Comparison of achieved and predicted tooth movement of maxillary first molars and central incisors: First premolar extraction treatment with Invisalign,” The Angle Orthodontist, vol. 89, no. 5, pp. 679–687, 2019, doi: 10.2319/090418-646.1.
[31] F. F. Dai, T. M. Xu, and G. Shu, “Comparison of achieved and predicted crown movement in adults after 4 first premolar extraction treatment with Invisalign,” American Journal of Orthodontics and Dentofacial Orthopedics, vol. 160, no. 6, pp. 805–813, 2021, doi: 10.1016/j.ajodo.2020.06.041.
[32] L. Ren et al., “The predictability of orthodontic tooth movements through clear aligner among first-premolar extraction patients: A multivariate analysis,” Progress in Orthodontics, vol. 23, no. 1, Art. no. 52, 2022, doi: 10.1186/s40510-022-00447-y.
[33] L. H. Ghislanzoni, Z. Kalemaj, M. Manuelli, C. Magni, A. Polimeni, and A. Lucchese, “How well does Invisalign ClinCheck predict actual results: A prospective study,” Orthodontics & Craniofacial Research, vol. 27, no. 3, pp. 465–473, 2024, doi: 10.1111/ocr.12752.
[34] T. Mourgues, M. J. González-Olmo, L. Huanca Ghislanzoni, C. Peñacoba, and M. Romero-Maroto, “Artificial intelligence in aesthetic dentistry: Is treatment with aligners clinically realistic?” Journal of Clinical Medicine, vol. 13, no. 20, Art. no. 6074, 2024, doi: 10.3390/jcm13206074.
[35] A. Polizzi, S. Serra, R. Leonardi, and G. Isola, “Clinical applications of artificial intelligence in teleorthodontics: A scoping review,” Medicina, vol. 61, no. 7, Art. no. 1141, 2025, doi: 10.3390/medicina61071141.
[36] J. F. McCray et al., “Accuracy of DentalMonitoring’s artificial intelligence in detecting aligner tracking issues: A retrospective multi-centric study,” BMC Oral Health, vol. 26, no. 1, Art. no. 239, 2026, doi: 10.1186/s12903-025-07580-0.
[37] J. Mongan, L. Moy, and C. E. Kahn Jr., “Checklist for artificial intelligence in medical imaging (CLAIM): A guide for authors and reviewers,” Radiology: Artificial Intelligence, vol. 2, no. 2, Art. no. e200029, 2020, doi: 10.1148/ryai.2020200029.
[38] F. Schwendicke et al., “Artificial intelligence in dental research: Checklist for authors, reviewers, readers,” Journal of Dentistry, vol. 107, Art. no. 103610, 2021.
[39] Align Technology, “Align Technology launches Align X-ray Insights, an AI computer-aided detection software, in the European Union and United Kingdom,” press release, Mar. 10, 2025.
