Retinal Age Model Tied to Disease Risk
Fundus-based tool estimates biologic aging but shows modest gains over prior models
By
Andrea Surnit
May 1, 2026
1
A deep-learning model estimated retinal age with a mean absolute error of about 3 years, indicating strong predictive performance.
2
Larger retinal age gaps were linked to cardiometabolic conditions, particularly in patients taking diabetes medication.
3
The model's performance declined in heterogeneous populations, highlighting the need for validation in diverse groups.
4
The study's observational design and reliance on self-reported data may limit the interpretability of its findings.
5
Further research is necessary to establish clinical thresholds and assess the model's impact on patient outcomes.
