arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2008.04370eess.IVcs.CV
收藏

使用深度学习预测糖尿病视网膜病变的风险

Predicting Risk of Developing Diabetic Retinopathy using Deep Learning

Google Health, Google LLC(谷歌有限责任公司谷歌健康);Advanced Clinical;Google Research, Google LLC(谷歌有限责任公司谷歌研究院);Adecco Staffing(德科 staffing);EyePACS LLC(EyePACS有限责任公司);Rajavithi Hospital(拉贾维蒂医院);College of Medicine, Rangsit University(兰实大学医学院)
Ashish Bora, Siva Balasubramanian, Boris Babenko, Sunny Virmani, Subhashini Venugopalan, Akinori Mitani, Guilherme de Oliveira Marinho, Jorge Cuadros, Paisan Ru… 展开作者

Ashish Bora, Siva Balasubramanian, Boris Babenko, Sunny Virmani, Subhashini Venugopalan, Akinori Mitani, Guilherme de Oliveira Marinho, Jorge Cuadros, Paisan Ruamviboonsuk, Greg S Corrado, Lily Peng, Dale R Webster, Avinash V Varadarajan, Naama Hammel, Yun Liu, Pinal Bavishi

摘要:

糖尿病视网膜病变(DR)筛查有助于预防失明,但随着糖尿病患者数量的增加,面临着规模化的挑战。DR发展的风险分层可能有助于优化筛查间隔,以降低成本,同时改善视力相关的结果。我们创建并验证了两个版本的深度学习系统(DLS),以预测接受DR筛查的糖尿病患者中轻度或重度(“轻度+”)DR的发展。这两个版本使用三个字段或单个字段的彩色眼底照片(CFPs)作为输入。训练集来自575,431只眼睛,其中28,899只已知2年结果,其余用于通过多任务学习来增强训练过程。对内部验证组(A组; 7,976只眼; 3,678只眼,结局已知)和外部验证组(B组; 4,762只眼; 2,345只眼,结局已知)进行验证。对于预测2年DR的发展,在验证集A上,3视野DLS的受试者工作特征曲线下面积(AUC)为0.79(95%CI,0.78-0.81)。在验证集B(仅包含单个视野)中,1个视野DLS的AUC为0.70(95%CI,0.67-0.74)。即使在调整了可用的危险因素后,DLS也是预后性的(p<0.001)。当添加到风险因素中时,3-视野DLS将验证集A中的AUC从0.72(95%CI,0.68-0.76)改善至0.81(95%CI,0.77-0.84),1-视野DLS将验证集B中的AUC从0.62(95%CI,0.58-0.66)改善至0.71(95%CI,0.68-0.75)。本研究中的DLSs从CFPs中确定了DR发展的预后信息。这些信息独立于现有的风险因素,并且比现有的风险因素提供更多的信息。

英文摘要:

Diabetic retinopathy (DR) screening is instrumental in preventing blindness, but faces a scaling challenge as the number of diabetic patients rises. Risk stratification for the development of DR may help optimize screening intervals to reduce costs while improving vision-related outcomes. We created and validated two versions of a deep learning system (DLS) to predict the development of mild-or-worse ("Mild+") DR in diabetic patients undergoing DR screening. The two versions used either three-fields or a single field of color fundus photographs (CFPs) as input. The training set was derived from 575,431 eyes, of which 28,899 had known 2-year outcome, and the remaining were used to augment the training process via multi-task learning. Validation was performed on both an internal validation set (set A; 7,976 eyes; 3,678 with known outcome) and an external validation set (set B; 4,762 eyes; 2,345 with known outcome). For predicting 2-year development of DR, the 3-field DLS had an area under the receiver operating characteristic curve (AUC) of 0.79 (95%CI, 0.78-0.81) on validation set A. On validation set B (which contained only a single field), the 1-field DLS's AUC was 0.70 (95%CI, 0.67-0.74). The DLS was prognostic even after adjusting for available risk factors (p<0.001). When added to the risk factors, the 3-field DLS improved the AUC from 0.72 (95%CI, 0.68-0.76) to 0.81 (95%CI, 0.77-0.84) in validation set A, and the 1-field DLS improved the AUC from 0.62 (95%CI, 0.58-0.66) to 0.71 (95%CI, 0.68-0.75) in validation set B. The DLSs in this study identified prognostic information for DR development from CFPs. This information is independent of and more informative than the available risk factors.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑