Google Health, Google, Mountain View, CA, USA.
Advanced Clinical, Deerfield, IL, USA.
Lancet Digit Health. 2021 Jan;3(1):e10-e19. doi: 10.1016/S2589-7500(20)30250-8. Epub 2020 Nov 26.
Diabetic retinopathy screening is instrumental to preventing blindness, but scaling up screening is challenging because of the increasing number of patients with all forms of diabetes. We aimed to create a deep-learning system to predict the risk of patients with diabetes developing diabetic retinopathy within 2 years.
We created and validated two versions of a deep-learning system to predict the development of diabetic retinopathy in patients with diabetes who had had teleretinal diabetic retinopathy screening in a primary care setting. The input for the two versions was either a set of three-field or one-field colour fundus photographs. Of the 575 431 eyes in the development set 28 899 had known outcomes, with the remaining 546 532 eyes used to augment the training process via multitask learning. Validation was done on one eye (selected at random) per patient from two datasets: an internal validation (from EyePACS, a teleretinal screening service in the USA) set of 3678 eyes with known outcomes and an external validation (from Thailand) set of 2345 eyes with known outcomes.
The three-field deep-learning system had an area under the receiver operating characteristic curve (AUC) of 0·79 (95% CI 0·77-0·81) in the internal validation set. Assessment of the external validation set-which contained only one-field colour fundus photographs-with the one-field deep-learning system gave an AUC of 0·70 (0·67-0·74). In the internal validation set, the AUC of available risk factors was 0·72 (0·68-0·76), which improved to 0·81 (0·77-0·84) after combining the deep-learning system with these risk factors (p<0·0001). In the external validation set, the corresponding AUC improved from 0·62 (0·58-0·66) to 0·71 (0·68-0·75; p<0·0001) following the addition of the deep-learning system to available risk factors.
The deep-learning systems predicted diabetic retinopathy development using colour fundus photographs, and the systems were independent of and more informative than available risk factors. Such a risk stratification tool might help to optimise screening intervals to reduce costs while improving vision-related outcomes.
Google.
糖尿病视网膜病变筛查对于预防失明至关重要，但由于各种类型糖尿病患者数量的增加，扩大筛查范围具有挑战性。我们旨在开发一种深度学习系统，以预测糖尿病患者在 2 年内发生糖尿病视网膜病变的风险。
我们创建并验证了两种深度学习系统版本，以预测在初级保健环境中进行远程视网膜糖尿病筛查的糖尿病患者发生糖尿病视网膜病变的发展。两个版本的输入均为一组三张或一张彩色眼底照片。在开发集的 575431 只眼中，有 28899 只已知结局，其余 546532 只用于通过多任务学习来增强训练过程。验证是在来自两个数据集的每个患者的一只眼（随机选择）上进行的：一个内部验证集（来自美国的远程视网膜筛查服务 EyePACS），其中包含 3678 只已知结局的眼，以及一个外部验证集（来自泰国），其中包含 2345 只已知结局的眼。
在内部验证集中，三张眼底照片深度学习系统的受试者工作特征曲线下面积（AUC）为 0.79（95%CI 0.77-0.81）。使用单张眼底照片深度学习系统评估仅包含单张眼底彩色照片的外部验证集，其 AUC 为 0.70（0.67-0.74）。在内部验证集中，现有风险因素的 AUC 为 0.72（0.68-0.76），将深度学习系统与这些风险因素结合后，AUC 提高至 0.81（0.77-0.84）（p<0.0001）。在外部验证集中，在将深度学习系统添加到现有风险因素后，AUC 从 0.62（0.58-0.66）提高至 0.71（0.68-0.75）（p<0.0001）。
深度学习系统使用眼底彩色照片预测糖尿病视网膜病变的发展，该系统独立于且比现有风险因素更具信息量。这种风险分层工具可能有助于优化筛查间隔，在提高视力相关结局的同时降低成本。
谷歌。
