Assessing Diagnostic Accuracy in Cephalometry: A Comparative Study of Manual and Digital Tracing Techniques

This study aimed to statistically evaluate and compare the accuracy, reliability, and efficiency of manual versus artificial intelligence (AI)-assisted digital cephalometric tracing using Steiner's and Down's analyses in orthodontic diagnostics.

A retrospective study was conducted using 20 lateral cephalograms obtained using the NewTom GiANO HR cone-beam computed tomography (CBCT) system (Quantitative Radiology, Verona, Italy). Manual tracings were performed on acetate sheets, while digital analysis employed the AudaxCeph® software (Audax d.o.o., Ljubljana, Slovenia) with automated landmark detection. Measurements were analysed using independent sample t-tests and Mann-Whitney U tests, with significance set at p ≤ 0.05.

Statistical analysis showed no significant differences (p > 0.05) between manual and digital cephalometric measurements across key parameters. Minor variations were within clinically acceptable limits, confirming that both methods are consistent and interchangeable for orthodontic diagnostics.

AI-assisted digital cephalometric tracing is as accurate and reliable as manual methods, offering enhanced efficiency and consistency. These findings support its integration into routine orthodontic diagnostics.