Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-1450
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-10-15T18:12:12.322916Z
status: published
lens: trace
sector: health
headline: Development and external validation of an AI pipeline for automated estimation of a radiographic DMFT (rDMFT) score from panoramic radiographs: a feasibility study
dek: 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…
gain_title: 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.
problem_title: 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.
trace_subject: Automated rDMFT estimation from panoramic radiographs
gain_reading: 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.
gain_evidence: the pipeline reached an ICC of 0.84 (95% CI 0.74 to 0.90) and an MAE of 2.38 | tooth loss was no longer overlooked (26 of 26 detected)
problem_reading: 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.
problem_evidence: the decayed component is unreliable | missing teeth were over-called in 15 of 74 complete dentitions
quick_read: Researchers combined a tooth-status detector and a tooth-numbering detector, trained on the DENTEX 2023 dataset, into a pipeline that reads one tooth per position on a panoramic radiograph. They tested it on 100 radiographs from an independent hospital against a reference set by two dentists working from the images.

The result suggests that automated radiographic dental scoring is feasible as a research tool, but the weak decayed-teeth component and imprecise missing-teeth counts show it is far from clinical use. The authors say any application would first need validation in unselected cohorts against clinically determined standards.
limitation: 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.
tag: Dual reading
key_points: Agreement with the reference was ICC 0.84 and MAE 2.38, compared with ICC 0.88 and MAE 2.13 between the two dentists. | Agreement was strongest for filled teeth (r 0.90) and weakest for decayed teeth (r 0.34, versus 0.16 between dentists). | Among 26 radiographs with missing teeth, the missing count was exact in 42% and within one tooth in 62%. | Adding tooth numbering raised detection of tooth loss to 26 of 26 but lowered overall ICC from 0.90 to 0.84.
rundown: The pipeline excluded third molars and computed missing teeth as 28 minus the number of occupied FDI positions. It was trained on DENTEX 2023 and validated externally on 100 radiographs from a different hospital.

Against each dentist individually the errors were larger, with MAE of 3.00 and 2.73. The pipeline also showed a positive bias of 1.14 units attributed to caries over-detection.

The authors describe the system as a research building block, not a decision-support or triage tool. They note rDMFT is a radiographic proxy rather than a clinical measure.
sources:
- peer_reviewed | Clinical Oral Investigations | https://doi.org/10.1007/s00784-026-07214-x | 2026-10-15
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

c4ec17632c3b4a784850b532ec5339408beaf6e6c7a266186347072adf674b0b

previous

0000000000000000000000000000000000000000000000000000000000000000