EYENET: An Eye Disease Detection System using Convolutional Neural Network

The eye is the most crucial sense organ for seeing the outside world. Diagnosing eye disease is complex, fallible, and time-consuming. Early detection and treatment can minimize a patient's suffering and stop blindness. Therefore, a computerized detection system is required to identify eye problems using fundus images. Such a method is possible with deep learning algorithms with enhanced image classification characteristics. The EYENET model uses Deep Convolutional Neural Networks to detect eye disorders at an early stage. The proposed EYENET model uses a Convolutional Neural Network (CNN), a self-diagnosis method, to predict five distinct eye diseases Bulging Eyes, Crossed Eyes, Cataracts, Glaucoma, and Uveitis, with high prediction rates. The CNN architecture works effectively and predicts diseases more accurately. The Adam optimizer is used to optimize the proposed network. The evaluation metrics used to justify the EYENET model are Precision, Recall, Accuracy and Fl-Score. The EYENET model had the highest accuracy 92.3%. The proposed mechanism facilitates to reduce workload for doctors, quick identification of diseases, and ease of use.