Abstract

Abstract: Retinal diseases, including diabetic retinopathy, glaucoma, and age-related macular degeneration, are leading causes of visual impairment and blindness across the globe. Detecting these conditions at an early stage is vital for effective treatment and prevention of vision loss. In recent years, deep learning techniques have demonstrated significant potential in diagnosing diseases from medical images. This study introduces a new method for detecting retinal diseases using convolutional neural networks (CNNs) applied to retinal images. The proposed method uses a CNN model to analyze retinal images and categorize them as either normal or showing signs of disease. Data augmentation techniques are also employed to expand the size and variability of the training dataset. Our system achieves high sensitivity and specificity in detecting diseases, underscoring its potential for application in clinical settings.