Comparative assessment of diagnostic agreement between artificial intelligence and general practitioners in diabetic retinopathy screening using non-mydriatic fundus photography
AI-assisted eye screening matches specialist accuracy, potentially extending diabetic retinopathy detection to primary care and catching vision threats earlier.
This cross-sectional study of 500 T2DM patients found that an AI-based fundus photography system achieved near-perfect diagnostic agreement with ophthalmologists and outperformed GPs on sensitivity, suggesting supervised primary care deployment could significantly strengthen DR screening programs. Large-scale multicentre validation is recommended before widespread adoption.
What the study was
- Study design
- Cross-sectional validation study (n=500 T2DM primary care patients)
- Population
- Type 2 diabetes mellitus primary care patients (mean age 64y, 59% male, DR prevalence 11%)
- Sample size
- 500
- Category
- Diagnostics
- Maturity
- Validated
- Journal
- Primary care diabetes
Why it surfaced
Strong AI validation study in primary care setting; near-perfect diagnostic performance; practical roadmap for implementation; cost-effectiveness analysis and multicentre validation still needed.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.