Automatic Exudate Detection with a Naive Bayes Classifier

Abstract

� Abstract—Diabetic retinopathy is a major cause of vision loss for diabetic patients, but early detection of its symptoms and treatment can prevent blindness. Exudate is a key indicator of diabetic retinopathy that can potentially be automatically quantified. In this paper, which focuses on automatic exudate detection in images acquired through non-dilated pupils, we present a series of experiments on feature selection and exudate pixel classification using a naive Bayes classifier. We find that the best feature set is a combination of 6 features: the pixel intensity after preprocessing, the standard deviation of the preprocessed intensities in a window around the pixel, the pixel hue, the number of edge pixels in a window around the pixel, the ratio between the size of the pixel's intensity cluster and the optic disc, and the response at the pixel to a derivative of Gaussian filter. The overall sensitivity, specificity, precision, and accuracy are 93.38%, 98.14%, 47.51%, and 98.05%, respectively.