Hybrid Deep Learning Models for Diabetic Retinopathy Classification

Diabetic retinopathy is a complication of diabetes in the eye. This disease is caused by the damage of the blood vessels of the back of eye (i.e., retina). Unfortunately, diabetic retinopathy can cause several symptoms, the most serious of which is complete vision loss. Indeed, the detection of diabetic retinopathy is a time-consuming manual process that requires a qualified clinician to examine and evaluate digital color photographs of the retina’s fundus. Currently, several researches are looking to employ artificial intelligence techniques, especially the Deep Learning, to deal with this issue. In this paper, we study some hybrid models for diabetic retinopathy severity classification in distributed and non-distributed environments. The studied models perform two main tasks: deep feature extraction and then classification of diabetic retinopathy according to its severity. The models were trained and validated on a publicly available dataset of 80,000 images and they achieved an accuracy of 80.7%. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.