Age-related macular degeneration (AMD)Age-related Macular Degeneration (AMD) is the leading cause of legal blindness in developed countries. Based on clinical features, the disease can be classified into early, intermediate, and late stages. In some patients, AMD advances to the late vision-threatening stage slowly; in others, the disease progresses faster and may quickly lead to a loss of vision in one or both eyes. There is, therefore, a critical need to detect AMD severity accurately. As deep learningDeep learning has become a rapidly expanding field in imaging-based analysis of AMD, we provide a short overview of recent advances in deep learningDeep learning applied to AMD severity classification. Our aims are to: (1) give a brief introductionIntroduction to AMD; (2) indicate how deep learningDeep learning has been applied to fundus photographs for AMD detectionDetection and classification; (3) highlight the potential utility of deep learningDeep learning methodologies to assist and enhance clinical decision-makingClinical decision-making in patients with AMD; and (4) provide a starting point for researchers who are interested in this field by providing useful resources such as code and data.