Automated tooth numbering on panoramic radiographs versus cone-beam computed tomographs: a diagnostic accuracy study...

To assess the diagnostic accuracy of a commercial artificial intelligence system for automated tooth numbering on panoramic radiographs and cone-beam CT (CBCT) and to quantify case-level reliability.

In this retrospective single-center diagnostic accuracy study, consecutive patients who underwent both panoramic radiography and CBCT in 2024 were included. The index test was automated tooth numbering generated by Diagnocat using the Fédération Dentaire Internationale numbering scheme. The reference standard was modality-specific consensus of 2 experienced, blinded readers. Tooth-position performance metrics with 95% confidence intervals were estimated using patient-level cluster bootstrap. Case-level reliability was defined as the proportion of examinations with completely error-free numbering across all evaluated tooth positions.

The study analyzed 171 paired panoramic radiographs and CBCT examinations, evaluating 4900 tooth positions per modality (9800 total). Tooth-position performance was identical and high across modalities (per-tooth accuracy 99.29% for both; overall position-based accuracy 98.73% [95% CI 97.97-99.38]). At the case level, 148/171 (86.5%) panoramic radiographs and 155/171 (90.6%) cone-beam computed tomography examinations were error-free.

Despite high tooth-position metrics, approximately 1 in 10 to 1 in 7 examinations required at least one manual correction, demonstrating a gap between granular accuracy and case-level reliability.

Reporting case-level, error-free outputs alongside tooth-position metrics provides a more clinically meaningful estimate of reliability for automated tooth numbering and supports safer workflow implementation.