Convolution Neural Network for Diabetic Retinopathy Detection

Abstract

Diabetes-Retinopathy (DR) condition detection based on machine learning and image processing techniques makes use of the diabetic portion from the set of input images. Textural feature analysis is adopted for feature extraction. CNN is used to classify the extracted features. The execution of the proposed technique is carried out in MATLAB, and the analysis is based on the accuracy, sensitivity, specificity. In the light of analytic outcomes, it can be said that the introduced method performs better than the existing technique in terms of all the mentioned parameters.