A Deep Learning approach to detect Diabetic Retinopathy with CNN and ResNet

• Published in: 2023 International Conference on Advances in Computing, Communication and Applied Informatics (ACCAI) pp. 1 - 7
• Main Authors: L, Antony Rosewelt, E, Nivetha, M, Sakthi Priya, S, Blessy Carolin
• Format: Conference Proceeding
• Language: English
• Published: IEEE 25-05-2023
• Subjects: Analytical models; Classification; CNN; Computer architecture; Data Visualization; Deep learning; Diabetic retinopathy; Predictive models; Resnet; Training; Visualization
• DOI: 10.1109/ACCAI58221.2023.10200176

Diabetic Retinopathy (DR) is a disease that is caused by long term diabetes mellitus, which causes lesions on the retina that impact vision. If it is not detected early, it could lead to blindness Deep learning algorithms have yielded encouraging results for detecting DR in retinal pictures. The goal of this research is to investigate the feasibility of employing deep convolutional neural networks (CNNs) to detect DR. The work will concentrate on training CNNs to classify different phases of DR utilising large-scale datasets of retinal pictures. The performance of different deep learning models, including ResNet and Inception, will be evaluated, and the robustness of these models to variations in image quality and disease severity will be assessed. The proposed research has the potential to contribute to the development of an accurate and efficient diagnostic tool for the early detection of DR, which can aid in preventing vision loss in diabetic patients. The findings of this research can potentially improve the clinical management of DR and provide better outcomes for patients.