Retinal imaging-based oculomics: artificial intelligence as a tool in the diagnosis of cardiovascular and metabolic diseases

AI Generated Summary*

A PRISMA-guided review surveys artificial intelligence applied to fundus photography, OCT and OCTA for cardiovascular and metabolic disease, drawing on 37 papers from January 2019 to May 2024 (29 cardiovascular, 8 metabolic). Reported models reached AUCs of 0.71 to 0.87, sensitivity 71 to 89 percent and specificity 40 to 70 percent, though the conclusion cites AUCs up to 0.97 for coronary disease, stroke and infarction prediction. Examples include a fundus atherosclerosis score (AUC 0.713) predicting cardiovascular death after Framingham adjustment, and a retinopathy algorithm at 87 to 90 percent sensitivity and roughly 98 percent specificity. Standardization, cross-population validation and image privacy stay unresolved, and conventional cardiac testing is still called the gold standard.

*This summary was generated by AI and is published unedited. Oku does not alter these summaries. It may contain errors or omissions and is provided for general informational purposes only. Oku does not guarantee its accuracy, completeness, or reliability. For authoritative information, please refer to the original, peer-reviewed article.