RETINAL DISEASE DETECTION USING DEEP LEARNING TECHNIQUES
B. Maneesha, G. Madhu Latha, Department of Computer Science, Dr. K.V. Subba Reddy MBA Institutions, Kurnool
DOI: https://doi.org/10.63712/bpsrj-v2i1p001
ABSTRACT:
Retinal disorders are among the leading causes of visual impairment worldwide, making early and accurate diagnosis essential for preventing irreversible vision loss. Recent advancements in deep learning have enabled automated analysis of medical images with high precision. This study investigates the application of deep learning models for the detection and classification of retinal diseases using fundus images. Convolutional Neural Networks (CNNs) are employed to automatically extract discriminative features from retinal scans, eliminating the need for manual feature engineering. The proposed approach is trained on labeled retinal datasets containing both healthy and diseased images, including conditions such as diabetic retinopathy, glaucoma, and age-related macular degeneration. Experimental results demonstrate that deep learning models achieve superior performance in terms of accuracy, sensitivity, and specificity compared to traditional machine learning techniques. Additionally, the use of attention mechanisms improves interpretability by highlighting disease-relevant regions within retinal images. The findings confirm that deep learning-based retinal disease detection systems can support clinical decisionmaking, enhance screening efficiency, and facilitate early intervention, particularly in resource-limited healthcare settings