Paper Title
Prediction of Heart Disease with Retinal Image Using Machine Learning

Abstract
Cardio Vascular Disease (CVD) continues to be a leading cause of mortality globally, stressing the importance of early detection for effective prevention and intervention. Retinal image analysis has emerged as a non-invasive approach for CVD risk prediction due to the close association of the retina with systemic vascular health. We present a hybrid deep learning model that leverages CNNs for enhanced performance, with Recurrent Neural Networks (RNN) to predict CVD risk factors including hypertension, diabetes, and hyperlipidemia. The CNN component extracts fine retinal features such as vessel changes and microaneurysms, which are early indicators of cardiovascular pathology, while the RNN component captures sequential dependencies in patient data to improve predictive accuracy. Our hybrid model shows high performance in identifying high-risk individuals by detecting subtle, yet critical retinal changes associated with cardiovascular health. In addition, the model interpretability framework provides insights into the correlation between retinal features and cardiovascular risk, which contributes valuable information to clinical decision-making. the research demonstrates the effectiveness of deep learning applied to retinal images for early and personalized CVD risk prediction. Keywords - Cardiovascular Disease, Retinal Imaging, Deep Learning, Convolutional Neural Networks, Recurrent Neural Networks, Risk Prediction