Diabetic Retinopathy (DR) is a medical condition which damages the retina and can cause permanent blindness to the people who have been suffering from diabetes since the last 15 years or more. Early diagnosis of the disease can save the damage of retina and thus can save the vision. Therefore, many intelligent systems have been proposed for early detection of DR. Many Machine Learning (ML) techniques have been used to propose those intelligent systems for detection of DR. Case-based reasoning (CBR) is one of the popular artificial intelligence (AI) techniques for problem solving which uses past experiences to solve problems. Recently, CBR is much used in image classification. Therefore, this paper proposes an intelligent system named case-based expert system for diabetic retinopathy (CBESDR) for early detection of DR which uses CBR and takes retinal fundus images as input for image processing (IP), extracts features, and retrieves similar cases from the case base using Euclidean distance(ED) similarity measure, and reuse and revise them to classify. The classification has been categorized into five categories: 0-No DR, 1-Mild Non-Proliferative DR(NPDR), 2-Moderate NPDR, 3-Severe NPDR, and 4-Proliferative DR to determine the severity of the disease. From the experimental results, it is observed that the proposed CBESDR gives significant performance.
