Early Diabetic Retinopathy Detection using Deep CNNs on Fundus Images

Author(s)	Mr. Yash Chiragkumar Shah, Mr. Krushang Bhavesh Tanti, Dr. Yogesh Kumar
Country	India
Abstract	Diabetic Retinopathy (DR) is a retinal disease caused by diabetes that, if not diagnosed early, can cause blindness. In this study, we present a novel deep learning technique to detect and classify diabetic retinopathy through retinal fundus images. In order to divide diabetic retinopathy into five categories: No_DR, Mild, Moderate, Severe and Proliferative_DR, six architectures CNN1D, CNN2D, VGG16, MobileNetV2, ResNet and DenseNet were used. According to the results of the work, DenseNet performed best in training by acquiring 84.39% training accuracy with a loss of 0.4092. At the same time, MobileNetV2 performed best in validation by acquiring 78.85% validation accuracy with a loss of 0.5923. DenseNet attained the most remarkable performance in class No_DR and Mild with classification F1-scores of 0.95 and 0.65 respectively. According to this work, AI based models are accurate in detecting and classifying the diabetic retinopathy. Moreover, it showcases the capabilities of DenseNet for early detection and clinical screening to be the most reliable and generalizable model.
Keywords	Diabetic Retinopathy, Deep Learning, CNN, DenseNet, MobileNetV2, Fundus Images, Classification
Field	Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In	Volume 8, Issue 2, March-April 2026
Published On	2026-04-12

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research