Ludwig Cassie A, Perera Chandrashan, Myung David, Greven Margaret A, Smith Stephen J, Chang Robert T, Leng Theodore
Department of Ophthalmology, Byers Eye Institute at Stanford, Stanford University School of Medicine, Palo Alto, CA, USA.
Retina Service, Department of Ophthalmology, Massachusetts Eye and Ear, Harvard Medical School, Boston, MA 02114, USA.
Transl Vis Sci Technol. 2020 Dec 4;9(2):60. doi: 10.1167/tvst.9.2.60. eCollection 2020 Dec.
To evaluate the performance of a deep learning algorithm in the detection of referral-warranted diabetic retinopathy (RDR) on low-resolution fundus images acquired with a smartphone and indirect ophthalmoscope lens adapter.
An automated deep learning algorithm trained on 92,364 traditional fundus camera images was tested on a dataset of smartphone fundus images from 103 eyes acquired from two previously published studies. Images were extracted from live video screenshots from fundus examinations using a commercially available lens adapter and exported as a screenshot from live video clips filmed at 1080p resolution. Each image was graded twice by a board-certified ophthalmologist and compared to the output of the algorithm, which classified each image as having RDR (moderate nonproliferative DR or worse) or no RDR.
In spite of the presence of multiple artifacts (lens glare, lens particulates/smudging, user hands over the objective lens) and low-resolution images achieved by users of various levels of medical training, the algorithm achieved a 0.89 (95% confidence interval [CI] 0.83-0.95) area under the curve with an 89% sensitivity (95% CI 81%-100%) and 83% specificity (95% CI 77%-89%) for detecting RDR on mobile phone acquired fundus photos.
The fully data-driven artificial intelligence-based grading algorithm herein can be used to screen fundus photos taken from mobile devices and identify with high reliability which cases should be referred to an ophthalmologist for further evaluation and treatment.
The implementation of this algorithm on a global basis could drastically reduce the rate of vision loss attributed to DR.
评估一种深度学习算法在检测通过智能手机和间接检眼镜镜头适配器获取的低分辨率眼底图像上转诊指征性糖尿病视网膜病变（RDR）的性能。
在92364张传统眼底相机图像上训练的自动深度学习算法，在来自两项先前发表研究的103只眼睛的智能手机眼底图像数据集上进行测试。图像从使用市售镜头适配器进行眼底检查的实时视频截图中提取，并作为以1080p分辨率拍摄的实时视频剪辑的截图导出。每位图像由一名获得委员会认证的眼科医生分级两次，并与算法的输出进行比较，该算法将每张图像分类为患有RDR（中度非增殖性DR或更严重）或无RDR。
尽管存在多种伪影（镜头眩光、镜头颗粒/污迹、使用者手部遮挡物镜）以及不同医学培训水平的使用者获得的低分辨率图像，但该算法在检测手机获取的眼底照片上的RDR时，曲线下面积为0.89（95%置信区间[CI]0.83 - 0.95），灵敏度为89%（95%CI 81% - 100%），特异性为83%（95%CI 77% - 89%）。
本文中完全基于数据驱动的人工智能分级算法可用于筛查从移动设备拍摄的眼底照片，并以高可靠性识别哪些病例应转诊给眼科医生进行进一步评估和治疗。
在全球范围内实施该算法可大幅降低因糖尿病视网膜病变导致的视力丧失率。
