| Affiliation |
(DM) (TYW) Singapore National Eye Centre, SIngapore, Singapore; (RJ) (CV) (DT) SERI, Singapore, Singapore; (SS) (JLL) (ST) SNEC, Singapore, Singapore; (ZJ) IHPC, AStar, Singapre, Singapore, Singapore; (MAF) Teheran Eye Institute, Teheran, Iran; (PF) Cooimbra eye Hospital, Coimbra, Portugal; (SH) Glostrup, Copehnagen University, Copenhagen, Denmark; (LM) Sunny Upstate Medical School, Syracuse; (NS) Geneva Hospital, Geneva, Switzerland; (CV) Fondation Rothschild, Paris, Paris, France; (CL) IRCCS Istituto delle Scienze Neurologiche di Bologna, Bologna, Italy; (MR) CHU Bordeaux, Bordeaux, France; (KV) Mahidol University, Bangkok, Thailand; (CC) Hong Kong University, Hong Kong, China; (JJ) Medical Faculty Mannheim of the Ruprecht-Karls-University of Heidelberg, Mannheim, Germany; (SA) Sankra Nathralya, Chennai, India; (PYWM) Cambridge, Moorfields, London, United Kingdom of Great Britain and Northern Ireland; (WL) University Freiburg Germany, Freiburg, Germany; (RK) Manila American Center, Manila, Philippines; (CF) The University of Sydney, Sydney, Australia; (NM)Johns Hopkins University School of Medicine, Baltimore, Maryland; (YL) IHPC, AStar, Singapore, Singapore; (JC) Department of Ophthalmology and Neurology, Mayo Clinic, Rochester, Rochester; (NN) Departments of Ophthalmology, Emory University School of Medicine, Atlanta, Atlanta; (VB) Emory University School of Medicine, Atlanta, Georgia |
| Description |
The timely and accurate detection of optic disc abnormalities, in particular papilledema, is a key part of the clinical examination; however, few physicians confidently perform ophthalmoscopy. Our goal was to develop an artificial intelligence deep learning system (AI-DLS) capable of distinguishing "normal optic nerves", "papilledema" (optic disc edema from proven intracranial hypertension), and "other optic nerve abnormalities" on standard digital fundus photographs from a large, multiethnic, worldwide, patient population. This classification was chosen to provide a low-cost, non-invasive test that facilitates identification of patients less (those with normal optic nerves) or more (those with papilledema) likely to have neurologic disorders associated with high morbidity/mortality in emergency settings. |
