Importance A deep learning system (DLS) is a machine learning technology with potential for screening diabetic retinopathy and related eye diseases. Objective To evaluate the performance of a DLS in detecting referable diabetic retinopathy, vision-threatening diabetic retinopathy, possible glaucoma, and age-related macular degeneration (AMD) in community and clinic-based multiethnic populations with diabetes. Design, Setting, and Participants Diagnostic performance of a DLS for diabetic retinopathy and related eye diseases was evaluated using 494 661 retinal images. A DLS was trained for detecting diabetic retinopathy (using 76 370 images), possible glaucoma (125 189 images), and AMD (72 610 images), and performance of DLS was evaluated for detecting diabetic retinopathy (using 112 648 images), possible glaucoma (71 896 images), and AMD (35 948 images). Training of the DLS was completed in May 2016, and validation of the DLS was completed in May 2017 for detection of referable diabetic retinopathy (moderate nonproliferative diabetic retinopathy or worse) and vision-threatening diabetic retinopathy (severe nonproliferative diabetic retinopathy or worse) using a primary validation data set in the Singapore National Diabetic Retinopathy Screening Program and 10 multiethnic cohorts with diabetes. Exposures Use of a deep learning system. Main Outcomes and Measures Area under the receiver operating characteristic curve (AUC) and sensitivity and specificity of the DLS with professional graders (retinal specialists, general ophthalmologists, trained graders, or optometrists) as the reference standard. Results In the primary validation dataset (n = 14 880 patients; 71 896 images; mean [SD] age, 60.2 [2.2] years; 54.6% men), the prevalence of referable diabetic retinopathy was 3.0%; vision-threatening diabetic retinopathy, 0.6%; possible glaucoma, 0.1%; and AMD, 2.5%. The AUC of the DLS for referable diabetic retinopathy was 0.936 (95% CI, 0.925-0.943), sensitivity was 90.5% (95% CI, 87.3%-93.0%), and specificity was 91.6% (95% CI, 91.0%-92.2%). For vision-threatening diabetic retinopathy, AUC was 0.958 (95% CI, 0.956-0.961), sensitivity was 100% (95% CI, 94.1%-100.0%), and specificity was 91.1% (95% CI, 90.7%-91.4%). For possible glaucoma, AUC was 0.942 (95% CI, 0.929-0.954), sensitivity was 96.4% (95% CI, 81.7%-99.9%), and specificity was 87.2% (95% CI, 86.8%-87.5%). For AMD, AUC was 0.931 (95% CI, 0.928-0.935), sensitivity was 93.2% (95% CI, 91.1%-99.8%), and specificity was 88.7% (95% CI, 88.3%-89.0%). For referable diabetic retinopathy in the 10 additional datasets, AUC range was 0.889 to 0.983 (n = 40 752 images). Conclusions and Relevance In this evaluation of retinal images from multiethnic cohorts of patients with diabetes, the DLS had high sensitivity and specificity for identifying diabetic retinopathy and related eye diseases. Further research is necessary to evaluate the applicability of the DLS in health care settings and the utility of the DLS to improve vision outcomes.

中文翻译：

---

使用来自多民族糖尿病人群的视网膜图像开发和验证糖尿病视网膜病变和相关眼病的深度学习系统

重要性 深度学习系统 (DLS) 是一种机器学习技术，具有筛查糖尿病视网膜病变和相关眼病的潜力。目的 评估 DLS 在检测社区和基于诊所的多民族糖尿病人群中可参考的糖尿病视网膜病变、威胁视力的糖尿病视网膜病变、可能的青光眼和年龄相关性黄斑变性 (AMD) 的性能。设计、设置和参与者使用 494 661 幅视网膜图像评估了 DLS 对糖尿病视网膜病变和相关眼病的诊断性能。训练 DLS 检测糖尿病视网膜病变（使用 76 370 张图像）、可能的青光眼（125 189 张图像）和 AMD（72 610 张图像），并评估 DLS 检测糖尿病视网膜病变（使用 112 648 张图像）、可能的青光眼的性能（71 896 张图片），和 AMD（35 948 张图像）。DLS 的培训于 2016 年 5 月完成，DLS 的验证于 2017 年 5 月完成，用于检测可参考的糖尿病视网膜病变（中度非增殖性糖尿病视网膜病变或更严重）和威胁视力的糖尿病视网膜病变（严重非增殖性糖尿病视网膜病变或更严重）使用新加坡国家糖尿病视网膜病变筛查计划和 10 个多民族糖尿病队列中的主要验证数据集。深度学习系统的使用。主要结果和措施 以专业分级师（视网膜专家、普通眼科医师、训练有素的分级师或验光师）为参考标准的受试者工作特征曲线 (AUC) 下的面积以及 DLS 的敏感性和特异性。结果 在主要验证数据集中（n = 14 880 名患者；71 896 张图像；平均 [SD] 年龄，60.2 [2.2] 岁；54.6% 男性），相关糖尿病视网膜病变的患病率为 3.0%；威胁视力的糖尿病视网膜病变，0.6%；可能的青光眼，0.1%；和 AMD，2.5%。DLS 对糖尿病视网膜病变的 AUC 为 0.936（95% CI，0.925-0.943），敏感性为 90.5%（95% CI，87.3%-93.0%），特异性为 91.6%（95% CI，91.0%-） 92.2%)。对于威胁视力的糖尿病视网膜病变，AUC 为 0.958（95% CI，0.956-0.961），敏感性为 100%（95% CI，94.1%-100.0%），特异性为 91.1%（95% CI，90.7%-91.4） %）。对于可能的青光眼，AUC 为 0.942（95% CI，0.929-0.954），敏感性为 96.4%（95% CI，81.7%-99.9%），特异性为 87.2%（95% CI，86.8%-87.5%）。对于 AMD，AUC 为 0.931（95% CI，0.928-0.935），敏感性为 93.2%（95% CI，91.1%-99.8%），特异性为 88.7%（95% CI，88.3%-89.0%）。对于 10 个额外数据集中可参考的糖尿病视网膜病变，AUC 范围为 0.889 至 0.983（n = 40 752 张图像）。结论和相关性 在对来自多民族糖尿病患者队列的视网膜图像的评估中，DLS 对识别糖尿病视网膜病变和相关眼病具有高度敏感性和特异性。需要进一步研究来评估 DLS 在医疗保健环境中的适用性以及 DLS 在改善视力结果方面的效用。