A Deep Learning System for Automated Angle-Closure Detection in Anterior Segment Optical Coherence Tomography Images
A Deep Learning System for Automated Angle-Closure Detection in Anterior Segment Optical Coherence Tomography Images
复制标题
DOI：
10.1016/j.ajo.2019.02.028
复制
发表时间：
2019-07-01
影响因子：
4.2
通讯作者：
Aung, Tin
中科院分区：
文献类型：
--
作者：
Fu, Huazhu;Baskaran, Mani;Aung, Tin
目的：前段光学相干断层扫描（AS-OCT）为视觉识别前段结构提供了客观的成像方式。一种自动检测系统可以帮助眼科医生解释AS-OCT图像中是否存在闭角。设计：开发一种人工智能闭角自动检测系统。方法：开发了一种用于AS-OCT图像自动闭角检测的深度学习系统，并将其与另一种基于定量特征的自动闭角检测系统进行比较。共检查了2113名受试者的4135张Visante AS-OCT图像（8270张前房角图像，其中7375张为开角，895张为闭角）。采用5次交叉验证对深度学习两类分类问题的闭角检测系统进行了测试。根据临床医生对as - oct图像的评分作为参考标准，对深度学习系统和基于定量特征的自动闭角检测系统进行评估。结果:采用定量特征的系统接收者工作特征曲线下面积为0.90(95%可信区间[CI] 0.891-0.914)，灵敏度为0.79 +/- 0.037，特异性为0.87 +/- 0.009，而深度学习系统接收者工作特征曲线下面积为0.96 (95% CI 0.953-0.968)，灵敏度为0.90 +/- 0.02，特异性为0.92 +/- 0.008，以临床医生对as - oct图像的分级为参考标准。结论：这些结果证明了深度学习系统在AS-OCT图像中进行闭合角检测的潜力。(C) 2019作者。Elsevier Inc.出版。
PURPOSE: Anterior segment optical coherence tomography (AS-OCT) provides an objective imaging modality for visually identifying anterior segment structures. An automated detection system could assist ophthalmologists in interpreting AS-OCT images for the presence of angle closure,DESIGN: Development of an artificial intelligence automated detection system for the presence of angle closure.METHODS: A deep learning system for automated angle closure detection in AS-OCT images was developed, and this was compared with another automated angle-closure detection system based on quantitative features. A total of 4135 Visante AS-OCT images from 2113 subjects (8270 anterior chamber angle images with 7375 open angle and 895 angle-closure) were examined. The deep learning angle-closure detection system for a 2-class classification problem was tested by 5-fold cross-validation. The deep learning system and the automated angle-closure detection system based on quantitative features were evaluated against clinicians' grading of AS-OCT images as the reference standard.RESULTS: The area under the receiver operating characteristic curve of the system using quantitative features was 0.90 (95% confidence interval [CI] 0.891-0.914) with a sensitivity of 0.79 +/- 0.037 and a specificity of 0.87 +/- 0.009, while the area under the receiver operating characteristic curve of the deep learning system was 0.96 (95% CI 0.953-0.968) with a sensitivity of 0.90 +/- 0.02 and a specificity of 0.92 +/- 0.008, against clinicians' grading of AS-OCT images as the reference standard.CONCLUSIONS: The results demonstrate the potential of the deep learning system for angle-closure detection in AS-OCT images. (C) 2019 The Authors. Published by Elsevier Inc.