Volume 30, Issue 174, August 2026

Artificial intelligence in bone age assessment for orthodontic treatment planning: a review of current evidence

Erwin Górski^1♦, Amelia Sobanek¹, Filip Szewczyk¹, Jakub Deptuch¹, Weronika Kubiak¹, Aleksandra Wyciślok¹, Justyna Koziarska¹, Aleksandra Pietrzyk², Zuzanna Szczerba³

¹Department of Orthodontics, Medical University of Lodz, 251 Pomorska Street, 92-216 Lodz, Poland
²Centralna Wojskowa Przychodnia Lekarska CePeLek SPZOZ, 78 Koszykowa Street, 00-911 Warszawa, Poland
³Navia Clinic, Dowborczyków 34, 90-019 Łódź, Poland
⁴Independent Public Healthcare Institution in Tarczyn, Warszawska 42, 05-555 Tarczyn, Poland

^♦Corresponding author
Erwin Górski, Institute of Dentistry of the Central Clinical Hospital of the Medical University of Łódź, Pomorska 251, 92-213 Łódź, Poland

ABSTRACT

Background: The accurate timing of orthodontic intervention depends primarily on assessment of a patient's residual growth potential. Conventional diagnostic approaches, such as cervical vertebral maturation (CVM) analysis and hand-wrist maturation (HWM) assessment, are subjective methods that rely solely on the clinician's visual interpretation. This study aimed to determine the extent to which artificial intelligence (AI), including deep learning (DL) algorithms, can be used to automate and standardise bone age assessment in orthodontic treatment planning. Methods: A structured narrative review was conducted. PubMed, Web of Science, Scopus, and Google Scholar databases were searched for articles published from January 2021 to March 2026. For the final analysis, 40 articles were selected. Results: Research shows that neural networks, such as convolutional neural networks (CNNs), can very well recognize the maturation stages of the cervical vertebrae and the bones of the hand and wrist. In selected studies, their accuracy exceeded 90%. Explainable AI, 3D CBCT analysis, and systems that combine multiple data types can improve the clarity, reproducibility, and clinical utility of such tools. Conclusions: Artificial intelligence may be helpful in decision-making and thus reducing diagnostic time and supporting the standardization of orthodontic treatment planning. Future research on multimodal systems, data protection, and transparent decision-making is needed.

Keywords: artificial intelligence; orthodontics; cervical vertebral maturation; bone age assessment; skeletal maturity

Medical Science, 2026, 30, e144ms3943

PDF

Published: 09 August 2026