aceDevelopment and external validation of an AI pipeline for automated estimation of a radiographic DMFT (rDMFT) score from panoramic radiographs: a feasibility study
Objectives: To assess the feasibility of estimating a radiographic Decayed-Missing-Filled-Teeth score (rDMFT), a proxy for the clinical DMFT index, from a single panoramic radiograph with a position-aware deep-learning pipeline. Materials and methods: A tooth-status detector and a tooth-numbering detector (32 FDI positions), trained on DENTEX 2023, were combined in a position-aware read-out (one tooth per FDI position, third molars excluded, missing teeth = 28 - occupied positions) and validated on 100 panoramic…
A position-aware deep-learning pipeline estimated a radiographic DMFT score from single panoramic radiographs with good agreement against a two-dentist reference, and it detected all tooth loss cases.
The same pipeline agreed less well than two dentists agree with each other, over-detected caries, over-called missing teeth in complete dentitions, and was unreliable for the decayed component.
The rDMFT is only a radiographic proxy, validated on 100 radiographs against a dentist reference from the images rather than clinical examination, and the authors say it is not proposed for clinical use.
Evidence
- Peer-reviewedClinical Oral Investigations2026-10-15
How should this claim be treated?
Truvace Impact Record TRV-2026-1450, v1: “Development and external validation of an AI pipeline for automated estimation of a radiographic DMFT (rDMFT) score from panoramic radiographs: a feasibility study.” Truvace, 2026-10-15. /record/TRV-2026-1450 (accessed at citation time). sha256 c4ec17632c3b4a78…
Calibration history
Every change to this record since certification, in the open. None yet — the reading has held since it entered the record.
Certified into the record
How to verify without trusting this page
Fetch the canonical text of any version from /api/record/TRV-2026-1450 and hash it yourself — for example shasum -a 256 on the saved canonical field. The result must equal content_hash, and each version’s text ends with prev:followed by the prior version’s hash (version 1 chains to 64 zeros). If a single character of any version had been altered since certification, the chain would not reproduce.