Abstract:Retinal diseases(mainly include retinal vascular diseases, extraretinal diseases such as outer retina, retinal pigment epithelium and subchoroidal diseases)are the leading causes of visual impairment and blindness worldwide, which affect human health and quality of life severely. In recent years, artificial intelligence(AI)technology, especially the applications of deep learning model are widespread. Innovations and new tools such as convolutional neural networks(CNNs), generative adversarial networks(GANs), Transformer architectures, show outstanding application value in early diagnosis, precise treatment, training and learning of ophthalmic diseases. Besides this, multimodal fusion models provide new ideas and tools for full-cycle management of ophthalmic diseases and related systemic diseases. This review aims to explore the application of AI or deep learning in the diagnosis of retinal diseases, and to discuss the current research status, progress, challenges and developments in future.
