Abstract
Purpose :
Atherosclerotic Cardiovascular Disease (ASCVD) remains a significant global health concern necessitating innovative approaches for early risk identification. This study aimed to validate Toku’s CLAiR technology prospectively and externally in predicting 10-year ASCVD risk using retinal images in the Middle Eastern population
Methods :
Our study recruited consented participants from the Moorfields Eye Hospital in Abu Dhabi, United Arab Emirates who came in for regular eye clinic examinations. Enrollment criteria included good, artifact-free retinal images and availability of basic medical and laboratory examination. Subjects with retinal disease were excluded. The retinal photos were taken with the TRC NW-400 (Topcon Healthcare, Tokyo, Japan) fundus camera. Age, gender, blood pressure, lipid panel and HbA1c were obtained from the electronic health records. These parameters were used to calculate the pooled cohort equation (PCE) score for these participants. The efficacy of CLAiR was then assessed by identifying people with elevated ASCVD risk (> 7.5% according to the American Heart Association) via retinal image vs traditional Pooled Cohort Equation (PCE) calculator. The study was approved by the Department of Health Ethics Committee
Results :
The study included 600 eyes of 300 patients with mean age of 43.5 ± 14.8 and 54% females. Early results (not including the entire dataset) achieved the accuracy of 89%, and sensitivity and specificity of 100% and 87.5% respectively, for detecting individuals with elevated ASCVD risk (PCE > 7.5%)
Conclusions :
This prospective study showcases the potential of CLAiR by using retinal images in non-invasively estimating ASCVD risk within our local population. The study not only contributes to the broader understanding of cardiovascular risk assessment in the Middle East (10.1% - 95% confidence interval: 7.1- 14.3%, p<0.001) but also holds implications for personalized patient care, aligning with the hospital's commitment to advancing healthcare through cutting-edge technology.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.