机构：[1]Casey Eye Institute, Oregon Health & Science University, Portland, OR, USA.[2]Department of Biomedical Engineering, Oregon Health & Science University, Portland, OR, USA.[3]Department of Ophthalmology, Aichi Medical University, Nagakute, Japan.[4]Shanxi Eye Hospital, Taiyuan, Shanxi, China.[5]State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou, China.[6]Beijing Institute of Ophthalmology, Beijing Tongren Eye Center, Beijing Tongren Hospital, Capital Medical University, Beijing Ophthalmology and Visual Science Key Lab, Beijing, China.研究所眼科研究所首都医科大学附属北京同仁医院首都医科大学附属同仁医院
To diagnose and segment choroidal neovascularization (CNV) in a real-world multicenter clinical OCT angiography (OCTA) data set using deep learning.A total of 105,66 OCTA scans from 3135 eyes, including 4701 with CNV and 5865 without, were collected in five eye clinics. Both 3 × 3-mm and 6 × 6-mm scans of the central and temporal macula were included. Scans with CNV were collected from multiple diseases, and scans without CNV were collected from both healthy controls and those with multiple diseases. No scans were removed during training or testing due to poor quality. The trained hybrid multitask convolutional neural network outputs a CNV diagnosis and membrane segmentation, respectively.The model demonstrated a highly accurate CNV diagnosis (area under receiver operating characteristic curve = 0.97), achieving a sensitivity of 95% at 95% specificity. The model also correctly segmented CNV lesions (F1 score = 0.78 ± 0.19). Additionally, model performance was comparable on both high-definition 3 × 3-mm scans and low-definition 6 × 6-mm scans. The model did not suffer large performance variations under different diseases. We also show that a subclinical lesion in a patient with neovascular age-related macular degeneration can be monitored over a multiyear time frame using our approach.The proposed method can accurately diagnose and segment CNV in a large real-world clinical data set.The algorithm could enable automated CNV screening and quantification in the clinic, which will help improve CNV diagnosis and treatment evaluation.
基金：
National Institutes of Health
(R01 EY024544, R01 EY027833, P30 EY010572,
T32 EY023211, UL1TR002369), an unrestricted
departmental funding grant and H. James and Carole
Free Catalyst Award from Research to Prevent Blindness (New York).
第一作者机构：[1]Casey Eye Institute, Oregon Health & Science University, Portland, OR, USA.[2]Department of Biomedical Engineering, Oregon Health & Science University, Portland, OR, USA.
通讯作者：
通讯机构：[1]Casey Eye Institute, Oregon Health & Science University, Portland, OR, USA.[2]Department of Biomedical Engineering, Oregon Health & Science University, Portland, OR, USA.[*1]Casey Eye Institute, Oregon Health & Science University, 515 SW Campus Dr., Portland, OR 97239, USA.
推荐引用方式(GB/T 7714)：
Wang Jie,Hormel Tristan T,Tsuboi Kotaro,et al.Deep Learning for Diagnosing and Segmenting Choroidal Neovascularization in OCT Angiography in a Large Real-World Data Set[J].TRANSLATIONAL VISION SCIENCE & TECHNOLOGY.2023,12(4):doi:10.1167/tvst.12.4.15.
APA：
Wang Jie,Hormel Tristan T,Tsuboi Kotaro,Wang Xiaogang,Ding Xiaoyan...&Jia Yali.(2023).Deep Learning for Diagnosing and Segmenting Choroidal Neovascularization in OCT Angiography in a Large Real-World Data Set.TRANSLATIONAL VISION SCIENCE & TECHNOLOGY,12,(4)
MLA：
Wang Jie,et al."Deep Learning for Diagnosing and Segmenting Choroidal Neovascularization in OCT Angiography in a Large Real-World Data Set".TRANSLATIONAL VISION SCIENCE & TECHNOLOGY 12..4(2023)
