CARDIORET-Cardiovascular Risk prediction using Retinal Eye Images

Heart disease is still among the main causes of death globally, necessitating accurate and timely detection techniques. Conventional diagnostic methods like blood tests, stress tests, and ECGs are frequently intrusive and time-consuming. Image processing is used to analyse retinal image features like vessel thickness, illumination, and morphology because retinal microvascular changes are closely linked to cardiovascular health. By linking ocular biomarkers with heart disease indicators, the model seeks to achieve high accuracy, providing a viable substitute for early detection and prevention. Effective and non-invasive diagnostic systems are essential given the increase in cardiovascular-related deaths. This project presents a model for determining the risk of a heart attack to analyse retinal eye images. Real time retinal eye images to identify characteristics that indicate vascular health. Heart disease is classified and its likelihood is predicted using the Random Forest (RF) and Kernel based Support Vector Machine (SVM) algorithms. The proposed system demonstrates how retinal imaging can serve as a predictive approach for cardiovascular risk assessment, supporting timely intervention and improved patient care.