International Research Journal of Education and Technology

Diabetic Retinopathy (DR) is a progressive retinal disorder caused by long-term diabetes and is one of the primary causes of preventable blindness worldwide. The disease damages retinal blood vessels, leading to microaneurysms, haemorrhages, exudates, and in severe cases, permanent vision loss. Early detection and accurate grading of DR are critical for effective treatment and prevention of complications. However, conventional screening methods rely on manual examination of retinal fundus images by ophthalmologists, which is time-consuming, resource-intensive, and subject to inter-observer variability. This creates a need for automated, efficient, and scalable diagnostic systems. Recent developments in Artificial Intelligence (AI) and Deep Learning have transformed medical image analysis, particularly through the use of Convolutional Neural Networks (CNNs). CNNs are capable of automatically extracting hierarchical features from retinal images without manual intervention, making them highly suitable for DR detection and classification. This study presents a comprehensive AI-based framework for diabetic retinopathy detection using CNN architectures. The system performance is evaluated using standard metrics such as accuracy, precision, recall, F1-score, and confusion matrix analysis. Experimental results demonstrate that the integration of preprocessing, data balancing, and optimized CNN architecture significantly improves detection accuracy and reduces false negatives. The findings confirm that AI-driven screening systems can support ophthalmologists by providing fast, reliable, and cost-effective diagnostic assistance. The proposed model contributes to the advancement of automated healthcare solutions and offers promising potential for large-scale diabetic retinopathy screening, particularly in resource-limited and rural settings.

Keywords: Diabetic Retinopathy, Artificial Intelligence, Convolutional Neural Networks