Advancements in Diabetic Retinopathy Detection: Innovations in Machine Learning and Deep Learning Techniques

Diabetes, a widespread health issue affecting people of all ages, leads to high blood sugar levels and can cause serious complications if unmanaged. One such complication is Diabetic Retinopathy (DR), which damages the retina's blood vessels, potentially resulting in vision problems or blindness. Artificial Intelligence (AI), including Machine Learning (ML) and Deep Learning (DL), is increasingly utilized to detect DR stages by analyzing medical images, particularly fundus images that capture the back of the eye and are crucial for DR diagnosis. This study explores the effectiveness of various AI algorithms, especially deep learning-based methods, in extracting features from fundus images for DR detection, classification, and segmentation. We identify limitations in current models and propose new approaches to reduce processing times and costs in medical diagnostics, making the technology more accessible and aiding in preserving vision for patients with DR. Our research is the first to investigate the impact of complexity on processing times and costs in healthcare applications using diverse deep learning techniques. We provide a comparative analysis of ML and DL methods focusing o