Predicting Stroke Risk Using Retinal Imaging with Artificial Intelligence: A Scoping Review of Current Evidence

Stroke remains a leading cause of mortality and long-term disability worldwide, with a substantial proportion of cases considered preventable through early risk identification and targeted interventions. Traditional stroke risk prediction models rely primarily on clinical and demographic variables and demonstrate only moderate predictive performance, often failing to capture subclinical vascular changes. In recent years, advances in retinal imaging and artificial intelligence (AI) have enabled the non-invasive assessment of microvascular and neurovascular alterations that may reflect underlying cerebrovascular risk. The retina shares anatomical and physiological characteristics with cerebral vasculature, making it a promising window for stroke risk assessment. As a growing body of literature explores AI-based analysis of retinal images for predicting stroke and related cerebrovascular outcomes, a comprehensive mapping of the existing evidence is needed. This scoping review protocol outlines the planned methods to systematically identify, characterize, and synthesize current research on the use of retinal imaging combined with artificial intelligence for stroke risk prediction.