Diabetic Retinopathy Detection and Grading using Deep learning

Introduction: One of the complications of diabetes disease is diabetic retinopathy (DR). Diabetic patients may suffer from total loss of sight that's if it is not detected and medicated early. The early detection of DR is very important during funds screening on regular basis. Detection and grading of DR are difficult because most fundus images suffer from undersaturation and noise.
Method: This paper proposes a new enhancement process as a solution to the poor quality of fundus images. It also proposes two architectures of convolutional neural network (CNN) models. The first one is the binary classifier of DR images into normal and abnormal. The second CNN architecture to classify severity grades of DR. In this study we also utilized different pre-trained convolutional neural network models to show the impact on the performance of the use of transfer learning from pre-trained CNN models vs newly defined architectures. The pre-trained CNN models and the two new proposed CNN models are tested using Messidor1, Messidor2, and Kaggle EyePACS datasets.
Results: The proposed binary classifier model results in F1-score 0.9387, 0.9629, and 0.9430 on Messidor-1, Messidor-2, and EyePACS datasets respectively. The proposed second model classifies the five grades with F1-score of 0.9133, 0.9226, and 0.9393 on Messidor1, Messidor2, and Kaggle EyePACS datasets respectively.
Conclusion: The new proposed CNN model proved its reliability and efficiency to detect DR and classify severity grades of DR in funds image. Preprocessing techniques enhanced the performance by 10.83% of accuracy and 0.13037 in AUC using the binary model.