What happened
Beijing Yuyi Ganlan Information Technology built DC-OPGAI-01, software that uses deep learning to identify teeth with suspected caries on panoramic radiographs of adults. The Center for Medical Device Evaluation (CMDE) of China's National Medical Products Administration reviewed the company's application for Class III registration. The trial was the maker's own, a retrospective randomized crossover multi-reader multi-case study of 402 subjects at three sites. Each of 12 dentists read every case with and without the software. Tooth-level sensitivity for caries was 82.40% with the software and 67.43% without, a difference of 14.97% (95% CI 7.28% to 22.66%). Specificity was 97.37% against 97.16%, which met the non-inferiority margin. Reading took 50.3677 seconds per case with the software and 65.1161 seconds without. On 13 February 2025, CMDE concluded that “该产品上市带来的受益大于风险” (the benefits of putting the product on the market outweigh the risks) and recommended approval. The dentists read stored images, so the software guided no care. The report does not state the reference standard for caries.
Sources
-
Regulatory document, by authors with no tie to the system.
组间差值为 14.97%,95%置信区间为(7.28%,22.66%),下限优于优效界值 0,优效性假设成立
经综合评价,认为该产品上市带来的受益大于风险,综合剩余风险可接受。
Details
AI system
- Reported configuration
-
Release 1 (full version 1.0.0)
- Workflow
-
The software analyzed panoramic radiographs with deep learning and marked the positions of teeth with suspected caries. Twelve dentists read 402 stored cases with and without the marks in a randomized crossover design.
Follow-up
- Conditions for replication
-
Patients are adults aged 18 or over. Trained dental clinicians make the diagnosis and do not base a clinical decision on the software alone.
Entry details
- Sector
-
Healthcare.
- Purpose
-
Beijing Yuyi Ganlan Information Technology built DC-OPGAI-01 to identify teeth with suspected caries on panoramic radiographs of adults.
29 more questions not yet checked
model, provider, role, configuration, inputs and outputs, autonomy, human oversight, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.
Cite this record
Peter H. Vartanian. (2026). DC-OPGAI-01 helped dentists find more tooth decay on X-rays (AIBD-0962). Artificial Intelligence Benefits Database, edition of 1 November 2026. https://petervartanian.xyz/aibd/records/AIBD-0962.html (archived at https://doi.org/10.7910/DVN/AADSII, version 7.0)