On April 5, the results of 2018 Diabetic Retinopathy: Segmentation and Grading Challenge (IDRiD) held by The IEEE International Symposium on Biomedical Imaging (ISBI) were announced. DeepDR team represented by associate Prof. Sheng Bin won champions in optic disc detection and central macular detection, the two most difficult projects. This marks a research breakthrough of SJTU in medical imaging and reflects the strength of our department.  

ISBI 2018 is one of the three top academic conference in the field of biomedical engineering. Diabetic retinopathy is a common blinding disease caused by diabetes with high morbidity. Early detection and treatment can lower the risk of visual loss significantly. Therefore, the early detection and treatment of diabetic retinopathy became a research hotspot of AI+medical treatment. IDRiD held by ISBI attracted dozens of universities, colleges and research institutes around the world including University of North Carolina, National University of Singapore, IFLYTEK, Peking University, Zhejiang University and Samsung Research Institute. There are three sub-projects, that is, focus segmentation, disease staging, optic disc and central macular detection. DeepDR team was the only team that won two champions.