Deep-learning retinal vessel calibre measurements and risk of cognitive decline and dementia

Previous studies have explored the associations of retinal vessel calibre, measured from retinal photographs or fundus images using semi-automated computer programs, with cognitive impairment and dementia, supporting the concept that retinal blood vessels reflect microvascular changes in the brain. Recently, artificial intelligence deep-learning algorithms have been developed for the fully automated assessment of retinal vessel calibres. Therefore, we aimed to determine whether deep-learning-based retinal vessel calibre measurements are predictive of risk of cognitive decline and dementia. We conducted a prospective study recruiting participants from memory clinics at the National University Hospital and St. Luke's Hospital in Singapore; all participants had comprehensive clinical and neuropsychological examinations at baseline and annually for up to 5 years. Fully automated measurements of retinal arteriolar and venular calibres from retinal fundus images were estimated using a deep-learning system. Cox regression models were then used to assess the relationship between baseline retinal vessel calibre and the risk of cognitive decline and developing dementia, adjusting for age, gender, ethnicity, education, cerebrovascular disease status, hypertension, hyperlipidemia, diabetes, and smoking. A total of 491 participants were included in this study, of whom 254 developed cognitive decline over 5 years. In multivariable models, narrower retinal arteriolar calibre (hazard ratio per standard deviation decrease = 1.258, <i>P</i> = 0.008) and wider retinal venular calibre (hazard ratio per standard deviation increase = 1.204, <i>P</i> = 0.037) were associated with increased risk of cognitive decline. Among participants with cognitive impairment but no dementia at baseline (<i>n</i> = 212), 44 progressed to have incident dementia; narrower retinal arteriolar calibre was also associated with incident dementia (hazard ratio per standard deviation decrease = 1.624, <i>P</i> = 0.021). In summary, deep-learning-based measurement of retinal vessel calibre was associated with risk of cognitive decline and dementia.