Explainable Deep Learning for Age-related Macular Degeneration Detection in OCT Images A Review

Age-related Macular Degeneration (AMD) is a leading cause of vision loss among older adults. Recent advancements in deep learning, particularly convolutional neural networks (CNNs), have significantly improved the accuracy of AMD detection using Optical Coherence Tomography (OCT) images. Despite their performance, these models often function as black boxes, limiting their trustworthiness in clinical environments. This review highlights key explainable artificial intelligence (XAI) techniques—including Grad-CAM, SHAP, and LIME—that have been utilized to interpret deep learning models in AMD classification tasks. We discuss their methodologies, applications, comparative strengths, and clinical relevance. The paper also identifies current challenges and suggests future directions for integrating interpretable AI into ophthalmic diagnostics.