Retinal Imaging Model Aids CAD Detection

Fundus-based model showed moderate-to-good accuracy and improved performance when combined with clinical risk factors in a high-risk cohort

1

A retinal imaging model showed moderate-to-good discrimination for detecting coronary artery disease (CAD) in a study of 417 patients.
2

The combined model achieved an AUROC of 0.802, outperforming models based on clinical risk factors alone or retinal data alone.
3

Quantitative retinal parameters like fractal dimension and vessel density were independently associated with CAD after adjustment.
4

The model's higher sensitivity suggests it may better identify patients at increased risk of CAD than rule out the disease.
5

Further validation in larger, diverse populations is needed before the retinal imaging model can be widely applied in clinical settings.

1