Machine Learning and Deep Learning Models Used to Detect Diabetic Retinopathy and Its Stages

Numerous medical research investigations have been conducted, and artificial intelligence (AI), machine learning (ML), and deep learning (DL) algorithms have grown in importance in recent years in terms of detecting and classifying diseases. In the field of explainable AI (XAI), methods are being developed to explain the predictions produced by AI systems. A large amount of data must be processed to forecast or classify diseases in the medical field. Massive volumes of data have been turned into useful information for decision-making using AI and DL methods. However, exploring the clear literature in the medical sector remains tough. Diabetic retinopathy (DR) is a devastating condition induced by diabetes mellitus (DM). Due to the progress of DM, a patient’s vision begins to deteriorate, leading to DR. This chapter discusses XAI as a method for AI-based systems to analyze and diagnose health data, and it presents a suggested strategy with the goal of establishing accountability and summarizes the research works in the medical area that employs various ML and DL approaches to detect non-proliferative DR and proliferative DR. A comparison of the evaluated methodologies is also discussed to gain a better knowledge of the current models. Performance analysis of microaneurysms (MA) and detailed segmentation methods of DR are also carried out.