Submit Manuscript  

Article Details


Deep Learning Techniques for Diabetic Retinopathy Detection

Author(s):

Sehrish Qummar, Fiaz Gul Khan*, Sajid Shah, Ahmad Khan, Ahmad Din and Jinfeng Gao  

Abstract:


Diabetes occurs due to the excess of glucose in the blood that may affect many organs of the body. The increase in blood sugar in the body causes many problems. One of the most prominent of these problems is Diabetic Retinopathy (DR). DR occurs due to the mutilation of the blood vessels in a retina. The detection of DR is complicated and time-consuming due to its features for the ophthalmologists. Therefore, automatic detection is required, recently different machine and deep learning techniques are being applied to detect and classify DR. In this paper, we conducted a study of the various techniques available in the literature for the identification/classification of DR, the datasets used, strengths and weaknesses of each method and provides the future directions. Moreover, we also discussed the different steps for the detection that are segmentation of blood vessels in a retina, detecting lesions and other abnormalities of DR in binary and multiclass classification.

Keywords:

Diabetic retinopathy, Deep learning, Convolutional Neural Network, Diabetes, Machine learning, Lesions Detection

Affiliation:

Department of Computer Science COMSATS University Islamabad, Abbottabad Campus , Department of Computer Science COMSATS University Islamabad, Abbottabad Campus , Department of Computer Science COMSATS University Islamabad, Abbottabad Campus , Department of Computer Science COMSATS University Islamabad, Abbottabad Campus , Department of Computer Science COMSATS University Islamabad, Abbottabad Campus , Department of Information Engineering, Huanghuai University, Henan



Full Text Inquiry