DEEP NEURAL NETWORK BASED APPROACH FOR RETINAL DISEASE DETECTION

This work focuses on the automated detection of retinal diseases such as Diabetic Retinopathy (DR) and Age-related Macular Degeneration (AMD) using MATLAB, GLCM (Gray Level Co-occurrence Matrix), and Deep Neural Networks (DNNs). Retinal images are processed through contrast enhancement, noise reduction, and segmentation techniques to extract meaningful features. GLCM is employed for texture-based feature extraction, while a deep learning model classifies the disease stages with high precision. To enhance practical usability, the system is integrated with hardware components such as Arduino, an LCD display, and a buzzer alert mechanism. The LCD screen displays the classification results, and the buzzer provides an alert if abnormalities are detected, ensuring immediate attention. This embedded approach makes the system suitable for real-time applications in hospitals, clinics, and remote healthcare centers. The project aims to offer an efficient, cost-effective, and accessible solution for early disease detection, potentially reducing vision loss through timely diagnosis and treatment.

Autorentext

B.Sridhar completed B.E. in Electronics and Communication Engineering, M.E. in Applied Electronics & Ph.D. in Information and Communication Engineering at Anna University. His area of interest includes Communication Systems and Image Processing. Currently he is working as a Associate Professor at SRM Valliammai Engineering College, Kattankulathur.