Deep Learning-Based Prediction of Cardiovascular Diseases from Retinal Images

Abstract

Cardiovascular diseases (CVDs) are a leading cause of mortality worldwide. Early detection and accurate diagnosis of CVDs are crucial for effective intervention and improved patient outcomes. Retinal imaging has emerged as a non- invasive and cost-effective technique for CVD prediction. This study aims to develop a deep learning model using convolutional neural networks (CNNs) and Mobile-Net architecture to predict CVDs from retinal images. The proposed model leverages the capabilities of CNNs to automatically learn relevant features from retinal images and Mobile-Net's lightweight design for efficient deployment. A large dataset of retinal images, including healthy individuals and CVD patients, is utilized for model training and evaluation. The retinal images are pre-processed, including resizing, normalization, and augmentation techniques, to enhance data quality and diversity. This model has the potential to support healthcare professionals in making informed decisions, enabling timely interventions and preventive healthcare strategies. Further validation and integration into clinical settings are warranted to fully assess its clinical utility and impact on patient care.