Published Paper Details:

Diabetic Retinopathy (DR) remains a leading cause of vision loss among diabetic patients globally, making its early detection and classification vital for effective treatment. Recent advancements in deep learning have shown significant promise in medical image analysis, particularly in ophthalmology. This paper proposes a comprehensive study on the application of transfer learning-based deep neural networks (DNNs) for the early detection and multi-stage classification of Diabetic Retinopathy using fundus images. Leveraging pre-trained convolutional neural networks such as VGG16, ResNet50, and InceptionV3, the model fine-tunes features to effectively classify the severity levels of DR, ranging from no DR to proliferative DR. The proposed approach is validated using benchmark datasets, achieving high accuracy, precision, and sensitivity, thereby demonstrating robustness in real-world clinical environments. The integration of transfer learning not only reduces the computational burden and training time but also enhances feature extraction capabilities, leading to improved diagnostic performance. This study emphasizes the potential of transfer learning-based DNNs as a reliable tool for ophthalmologists in the early detection and grading of Diabetic Retinopathy.