A Benchmark of Ocular Disease Intelligent Recognition: One Shot for Multi-disease Detection

In ophthalmology, early fundus screening is an economic and effective way to prevent blindness caused by ophthalmic diseases. Clinically, due the lack of medical resources, manual diagnosis time-consuming may delay condition. With development deep learning, some researches on diseases have achieved good results, however, most them are just based one disease. During screening, ophthalmologists usually give diagnoses multi-disease binocular image, so we release a dataset with 8 meet real scene, which contains 10,000 images from both eyes 5,000 patients. We did benchmark experiments it through state-of-the-art neural networks. found simply increasing scale network cannot bring results for classification, well-structured feature fusion method combines characteristics needed. Through this work, hope advance research related fields.