Cephalometric analysis is a vital diagnostic tool in orthodontics, yet manual tracing is labor-intensive and prone to observer variability. The adoption of artificial intelligence (AI), especially convolutional neural networks (CNNs), is revolutionizing this process through automation and increased consistency.

This review evaluates the accuracy, efficiency, and clinical applicability of AI-assisted cephalometric analysis in comparison with conventional manual and digital methods.

A comprehensive literature review was performed using databases including PubMed, Scopus, and Web of Science. Included studies (2013–2025) evaluated AI-based 2D and 3D cephalometric systems on landmark detection accuracy, tracing time, reproducibility, and clinical reliability.

AI-assisted cephalometric analysis demonstrates high accuracy, reproducibility, and efficiency. Nevertheless, limitations persist in detecting dental landmarks and generalizing across diverse populations. A hybrid human-AI approach remains optimal for clinical deployment.

Keywords: Artificial intelligence; Cephalometry; Orthodontics; Deep learning; CNN; Automated landmark detection; Diagnostic imaging.