What happened
Tencent Healthcare (Shenzhen) built MY-YD-01, software that analyzes a single non-mydriatic color fundus photograph with deep learning modules. They check image quality and segment the optic disc and cup. The software then flags suspected chronic glaucoma-like optic neuropathy and suggests whether the patient needs further examination. 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, with 1,092 subjects at four institutions. It compared a trial group with a control group, and the report does not describe how each group read the photographs. Subject-level sensitivity was higher in the trial group by 16.42% (95% CI 3.09% to 28.97%), and specificity met the non-inferiority test. The report adds that the software alone had fewer false positives and false negatives than either group. On 29 August 2022, CMDE judged that “该产品受益大于风险” (the product's benefits outweigh its risks) and recommended approval. It lists doctors' overreliance on the software as a risk, and the software is not for people with high myopia.
Sources
-
Regulatory document, by authors with no tie to the system.
受试者水平检出灵敏度,FAS 及 PPS 集的试验组与对照组差值分别为16.42% ( 95%CI : 3.09%-28.97% ) 和 18.97% ( 95%CI :3.60%-33.08%) ,均满足优效性验证。
根据申请人提供的申报资料,经综合评价,在目前认知水平上,认为该产品受益大于风险。
Details
AI system
- Reported configuration
-
Release V1
- Workflow
-
Deep learning modules checked image quality, segmented the optic disc and cup and classified each accepted photograph. The software fused the predictions of a disc-and-cup model and a whole-image model into one prompt.
Follow-up
- Conditions for replication
-
Photographs are single non-mydriatic color fundus images of about 45 degrees from a specified camera that pass the software's quality check. Patients are adults, and people with high myopia are excluded. Doctors qualified for fundus examination make the decision and do not rely on the prompt alone.
Entry details
- Sector
-
Healthcare.
- Purpose
-
Tencent Healthcare (Shenzhen) built MY-YD-01 to flag suspected chronic glaucoma-like optic neuropathy on fundus photographs of adults and suggest whether they need further examination.
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). MY-YD-01 raised detection of glaucoma-like damage on fundus photos (AIBD-1073). Artificial Intelligence Benefits Database, edition of 1 November 2026. https://petervartanian.xyz/aibd/records/AIBD-1073.html (archived at https://doi.org/10.7910/DVN/AADSII, version 7.0)