These authors contributed equally: Yiming Qin, Tao Yan, Matthew Yu Heng Wong.

School of Clinical Medicine, Beijing Tsinghua Changgung Hospital, Tsinghua Medicine, Tsinghua University, Beijing, China

Yiming Qin & Tien Yin Wong

Department of Automation, Tsinghua University, Beijing, China

Tao Yan & Jiamin Wu

School of Clinical Medicine, University of Cambridge, Cambridge, UK

Matthew Yu Heng Wong

Center for Innovation and Precision Eye Health and Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore

Sahana Srinivasan & Yih-Chung Tham

Singapore Eye Research Institute, Singapore National Eye Centre, Singapore, Singapore

Sahana Srinivasan, Yih-Chung Tham & Tien Yin Wong

School of Biomedical Engineering, Tsinghua University, Beijing, China

Hong-Yu Zhou

Digital Health Center, Berlin Institute of Health, Charité – Universitätsmedizin Berlin, Berlin, Germany

Roland Eils

Intelligent Medicine Institute, Fudan University, Shanghai, China

Roland Eils

Ophthalmology and Visual Science Academic Clinical Program, Duke-NUS Medical School, Singapore, Singapore

Yih-Chung Tham

Clinical Excellence Research Center, Stanford University, Stanford, CA, USA

Nigam H. Shah

Center for Biomedical Informatics Research, Stanford University, Stanford, CA, USA

Nigam H. Shah

First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, China

Haibo Wang

Beijing Visual Science and Translational Eye Research Institute (BERI), Beijing Tsinghua Changgung Hospital Eye Center, School of Clinical Medicine, Tsinghua University, Beijing, China

Tien Yin Wong

Beijing Key Laboratory of Intelligent Diagnostic Technology and Devices for Major Blinding Eye Diseases, Beijing, China

Tien Yin Wong

T.Y.W. conceived and supervised the project. Y.Q., T.Y. and M.Y.H.W. wrote the paper. S.S., H.-Y.Z., R.E., Y.-C.T., N.H.S., J.W., H.W. and T.Y.W. reviewed and edited the paper. All authors provided critical comments and reviewed the paper. All authors discussed the results and approved the final version before submission.