Diabetic Retinopathy Prediction Using Deep Learning Techniques
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
This paper presents the development of an intelligent healthcare system leveraging deep learning and machine learning models for the early detection of diabetic retinopathy (DR) severity and the prediction of kidney disease risk. The system uses EfficientNet-B3 and other CNN architectures to classify retinal fundus images into DR severity levels, supported by Random Forest classifiers for kidney risk prediction based on clinical parameters. Advanced techniques such as data augmentation, focal loss, and transfer learning are employed to enhance model performance. The solution ensures robust and accurate diagnostics, accessible through a userfriendly Streamlit web application for real-time predictions. Experimental results demonstrate high accuracy and usability, offering valuable support for healthcare professionals in early intervention and personalized patient management.
Fields of Study & Subjects
Retinal Imaging and Analysis
Artificial Intelligence in Healthcare
Retinal Diseases and Treatments
Publication details
- Journal / source
- International Conference on Intelligent Systems for Communication, IoT and Security (ICISCoIS)
- Pages
- 1–5
- Publication date
- 8 January 2026
- Language
- EN