Diagnosis of Diabetic Retinopathy Using Data Mining Classification Techniques

Abstract

Diabetic retinopathy is one of the complications of diabetes that affects the small vessels of the retina, being the main cause of blindness in adults. An early detection of this disease is essential, as it can prevent blindness as well as other irreversible harmful outcomes. This article attempts to develop a data mining model capable of identifying diabetic retinopathy in patients based on features extracted from eye fundus images. The data mining process was carried out in the RapidMiner software and followed the CRISP-DM methodology. In particular, classification models were built by combining different scenarios, algorithms, and sampling methods. The data mining model which performed best achieved an accuracy of 76.90%, a precision of 85.92%, and a sensitivity of 67.40%, using the Logistic Regression algorithm and Split Validation as the sampling method.