Region-Based Convolutional Neural Networks Based Automated Detection and Classification of Diabetic Retinopathy

Abstract - One of the main causes of vision impairment in diabetic patients is diabetic retinopathy (DR), which is brought on by long-term elevated blood sugar levels damaging the retinal blood vessels. Early and accurate detection is essential for effective treatment. Deep learning-based approaches have shown significant promise in automated Diabetic Retinopathy (DR) diagnosis by leveraging advanced feature extraction and classification techniques. This study explores a deep learning framework utilizing a Region-based Convolutional Neural Network (RCNN) for Diabetic Retinopathy (DR) detection and severity classification. Region-based Convolutional Neural Network (RCNN) is employed for robust feature extraction, effectively capturing complex patterns and localized lesions in retinal images. The proposed model processes retinal fundus images to detect abnormalities with high precision and classify Diabetic Retinopathy (DR) into different severity levels. The results indicate that the Region-based Convolutional Neural Network (RCNN) based approach enhances classification performance, providing a reliable and scalable solution for automated Diabetic Retinopathy (DR) screening. Key Words: Diabetic Retinopathy, Region-based Convolutional Neural Network, Random Forest, Data Sets, Performance Metrics.