Peer-Reviewed Publication
Asia Pac J Ophthalmol (Phila)2026;100379.September 25, 2026Journal Article
A multi-label deep learning system for simultaneous detection of nine fundus conditions.
Jungwoo Ha1, Yeon Hee Choi1, Danbi Lee2, Junseo Choi2, Wonyoung Seo3, Geunyoung Lee3, Dongjin Nam4, Sahil Thakur5, Tae Keun Yoo6, Sunjin Hwang7
1Department of Ophthalmology, Hanyang University Guri Hospital, Guri City, South Korea; Deparment of Ophthalmology, Hanyang University College of Medicine, Seoul, South Korea.
2Deparment of Ophthalmology, Hanyang University College of Medicine, Seoul, South Korea.
3Mediwhale Inc., Seoul, South Korea.
4Mediwhale Inc., Seoul, South Korea; Department of Internal Medicine, Severance Hospital, Yonsei University College of Medicine, Seoul, South Korea.
5Mediwhale Inc., Seoul, South Korea; Singapore Eye Research Institute, Singapore National Eye Centre, Singapore, Republic of Singapore.
6Department of Ophthalmology, Bright Saint Mary's Eye Center, Seoul, South Korea.
7Department of Ophthalmology, Hanyang University Guri Hospital, Guri City, South Korea; Deparment of Ophthalmology, Hanyang University College of Medicine, Seoul, South Korea. Electronic address: sunjin1989@hanyang.ac.kr.
Abstract
PURPOSE: To develop and validate a deep learning model for simultaneous detection of nine fundus conditions from a single color fundus photograph.
METHODS: A development dataset of 236,242 color fundus images was assembled from 17 heterogeneous sources and partitioned at the patient level into training (70%), tuning (10%), and internal validation (20%) sets; 5-fold cross-validation was applied on…
Create a free account to keep reading
Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.
Want unlimited research access? See Pro plans