Mao J, Luo Y, Liu L, et al. - In the present study, the researchers created an end-to-end automated system that can not only diagnose plus disease but also provide quantitative characterization of pathological characteristics. The deep learning network provided retinal vessel segmentation and the optic disc. The vessel segmentation categorized plus disease and automatically assessed tortuosity, width, fractal dimension, and vessel density. For the diagnosis of plus disease, the trained network achieved a sensitivity of 95.1% with 97.8% specificity. The sensitivity and specificity were 92.4% and 97.4% for detection of preplus or worse. Quantitative analysis exhibited significant changes in the pathological characteristics after treatment for plus disease with ranibizumab injection. The system accomplished high precision of diagnosis of plus disease in retinopathy of prematurity. It provided a quantitative analysis of the disease progression's dynamic features. This automated system can help specialists with an auxiliary quantitative assessment of the typical pathological characteristics of the disease by providing a classification decision.