Dry and Wet Age-Related Macular Degeneration Classification Using OCT Images and Deep Learning

Diabetic retinopathy (DR) and age-related macular degeneration (AMD) are diseases that can have adverse effects on the eyes of the elderly. They affect the central part of the retina, called macula. Depending on the severity, they might require urgent eye care and the treatment varies according to the specific case. In this paper, automated and fast classification of dry and wet AMD using deep convolutional neural networks is proposed. It is important that both dry and wet types are accurately detected for timely treatment. It is shown here through the performance results of the deep neural networks that dry vision impairment can be detected more accurately than wet. It is further shown that eighteen layer ResNet model outperforms AlexNet model in classifications. The area under the receiver operating characteristic curve of the ResNet model for each AMD stage is 94% and 63%, respectively.