Related Experiment Videos
Quantifying the Economic Value of Artificial Intelligence-Assisted Diabetic Retinopathy Screening: A Meta-Analysis
Chen Wei1, Liujin Zhang1, Chen Chen1
1Center for Health Policy and Health Economics, National Infrastructures for Translational Medicine, Institute of Clinical Medicine, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, Beijing 100730, China.
Summary
Artificial intelligence (AI) screening for diabetic retinopathy (DR) is cost-effective, offering significant economic benefits over human graders or no screening. This technology is particularly valuable for expanding DR screening in underserved regions.
Area of Science:
- Ophthalmology
- Health Economics
- Artificial Intelligence
Background:
- Diabetic retinopathy (DR) is a leading cause of preventable blindness globally.
- Early detection through screening is crucial for preventing vision loss.
- The economic value of AI-assisted DR screening requires quantitative synthesis.
Purpose of the Study:
- To quantitatively synthesize the economic value of AI-assisted screening for DR.
- To compare the cost-effectiveness of AI-assisted DR screening with traditional methods.
Main Methods:
- Systematic review and meta-analysis of model-based economic evaluations.
- Searched eight databases for studies from January 2015 to August 2025.
- Pooled incremental net monetary benefit (INMB) using a random-effects model.
Main Results:
- 14 studies were included, with 11 in the meta-analysis.
- AI-assisted DR screening was significantly more cost-effective than human grading (pooled INMB: $2,179.39 per individual).
- AI-assisted screening showed a pooled INMB of $3,606.10 per individual compared to no screening from a societal perspective.
Conclusions:
- AI-assisted DR screening is a cost-effective strategy, especially for resource-limited settings.
- Adoption in established programs should consider local factors like ophthalmologist costs.
- Future evaluations should use real-world evidence and standardized reporting.
