International Journal of Science, Engineering and TechnologyAuthors: Usha C R, Kartik Narayan Bhat, Naveen Bhavi, Nikhil Patil, S.M Abhishek
Abstract: The rise in global cases of vision-threatening conditions such as Diabetic Retinopathy (DR) and Glaucoma underscores the urgent need for early and accurate diagnosis. However, traditional diagnostic approaches are often limited by accessibility, cost, and the requirement for specialized personnel. This research presents a novel, deep learning-based framework that utilizes retinal fundus images to simultaneously detect Diabetic Retinopathy and Glaucoma. Leveraging the power of Deep Convolutional Neural Networks (DCNN), the system analyzes ocular features and pathological markers with high precision, enabling efficient, non-invasive screening. The proposed model incorporates advanced image processing techniques and pre-trained CNN architectures to extract deep visual features from fundus photographs. It is trained and validated on publicly available retinal image datasets, ensuring robustness and generalizability. Furthermore, the system includes a smart triage mechanism that classifies patient severity levels to facilitate timely medical intervention. By integrating Artificial Intelligence, Machine Learning, and Computer Vision, this work aims to improve early detection rates, reduce diagnostic latency, and support overburdened healthcare systems, particularly in resource-constrained regions. The dual-disease focus enhances its clinical utility and scalability, making it a promising solution in the evolving field of AI-assisted ophthalmology.
