To evaluate the accuracy of a deep learning model in estimating biological aging through retinal imaging and its implications for systemic disease screening.
Key Findings:
The model demonstrated improved accuracy over previous systems.
Patients with diabetes, cardiac disease, or a history of stroke had significantly higher retinal age gaps.
The model's predictions focused on the optic disc, macula, and major vascular arcades, indicating their relevance to systemic vascular health.
Interpretation:
Limitations:
The study population was predominantly Asian, affecting generalizability.
Performance declined in more diverse datasets.
Image quality and acquisition variability influenced accuracy.
