Visionary AI: Transforming Diabetic Retinopathy Discovery through Advanced Deep Learning Models

• Published in: International Conference on Computing Communication Control and Automation (Online) pp. 1 - 7
• Main Authors: Arun, Bhingardive Akshada, Bansode, B. N.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 23.08.2024
• Subjects: AI (Artificial Intelligence); Computer architecture; Confusion Matrix; Convolutional Neural Network(CNN); Convolutional neural networks; Deep learning; Diabetic retinopathy; DR (Diabetic Retinopathy); F1 score; Measurement; Pathology; Precision; Recall; Refining; Retina; Transfer Learning; Transforms; Visual impairment
• ISSN: 2771-1358
• DOI: 10.1109/ICCUBEA61740.2024.10774872

Leveraging AI for DR detection can transform patient outcomes by enabling early intervention and preventing vision loss. Diabetic Retinopathy (DR) is a significant contributor to visual impairment and blindness globally. It predominantly affects patients who have lived with diabetes for an extended duration. The work aims to develop an effective representation for retinal images, enhancing the recognition. This paper discusses the implementation of a transfer learning-based method for DR discovery with a deep Convolutional Neural Network (CNN) using Inception-v3, VGG-16, Alexnet, Googlenet and ResNet 18 architecture on the APTOS 2019 Dataset and metrics like the precession, confusion matrix, f1- score and recall evaluated. In summary, these studies underscore the transformative potential of deep learning in revolutionizing the judgement of DR.