Medical-AI trial: detecting periapical lesions from dental panoramic images with AI — catching easily missed root-tip lesions (CBCT reference) (NCT05888935)
An observational study evaluating whether AI can detect periapical lesions (at the tip of a tooth root) from low-dose dental panoramic X-rays. Panoramic is a first-line examination, but detection rates for these lesions are low (20–36%), and cone-beam CT (CBCT) is used as the reference. The primary outcome is the performance of the AI software. 2,000 patients; recruiting.
Trial overview (primary data)
- StatusRecruiting
- ConditionsPeriapical Diseases
- SponsorCentre Hospitalier Régional Metz-Thionville
- Target enrollment2,000 participants
- Period2022-10-01 〜 2027-12-01
Key points
- An observational study of whether AI can detect periapical (root-tip) lesions from low-dose dental panoramic X-rays.
- Panoramic is a first-line examination, but detection rates for these lesions are low (about 20% panoramic, 36% retro-alveolar).
- It uses the more precise three-dimensional cone-beam CT (CBCT) as the reference.
- The primary outcome is the performance of the AI software. Large at 2,000 patients; recruiting.
- Flagging lesions at the panoramic stage could reduce misses and triage CBCT more smartly; a performance evaluation, not a replacement.
At the tip of a tooth root, lesions (such as a pocket of pus) can form from causes such as decay or inflammation of the tooth nerve. On dental X-rays these appear as a dark, radiolucent shadow, but early ones or areas with overlap are easily missed.
Widely taken dental panoramic X-rays involve low radiation and are convenient, yet per the summary the detection rate for periapical lesions is low (about 20% for panoramic and 36% for retro-alveolar radiographs), and confident assessment often needs the higher-radiation three-dimensional cone-beam CT (CBCT).
This study examines whether AI can make up for those misses, as an observational study. Per the registry summary, it evaluates whether AI can detect periapical lesions from low-dose panoramic images, using the more precise CBCT as the reference. The primary outcome is the performance of the AI software. The size is large at 2,000 patients, and it is recruiting.
If AI can accurately flag lesions at the convenient, low-dose panoramic stage, it might reduce misses while reserving higher-radiation CBCT for cases that truly need it. This is one example of the practical value of imaging AI, raising accuracy at the entry examination to make triage to more invasive tests smarter. The scale of 2,000 also strengthens the reliability of the evaluation.
That said, this study evaluates AI detection performance; it does not establish replacement of CBCT or of dentist judgment.
Why it matters
If AI can flag lesions at the convenient, low-dose panoramic stage, it could reduce misses and reserve higher-radiation CBCT for cases that truly need it, one example of raising entry-examination accuracy to make triage smarter (this study evaluates detection performance, not a replacement).
FAQ
What is a periapical lesion?
Does AI replace CBCT or the dentist judgment?
Sources (primary)
Source: ClinicalTrials.gov (U.S. NIH/NLM, public domain). This site does not provide medical advice. Verify the latest and exact details with the official source. This site is not endorsed or certified by the NIH/NLM.
- ClinicalTrials.gov (study record, original)
- NCT ID: NCT05888935