Bratt Alex, Guenther Zachary, Hahn Lewis D, Kadoch Michael, Adams Patrick L, Leung Ann N C, Guo Haiwei H
Department of Radiology, Stanford University School of Medicine, 300 Pasteur Dr, Stanford, CA 94305 (A.B., Z.G., L.D.H., P.L.A., A.N.C.L., H.H.G.); and Department of Radiology, University of California at Davis, Sacramento, Calif (M.K.).
Radiol Cardiothorac Imaging. 2019 Dec 19;1(5):e190057. doi: 10.1148/ryct.2019190057. eCollection 2019 Dec.
To test the performance of a deep learning (DL) model in predicting atrial fibrillation (AF) at routine nongated chest CT.
A retrospective derivation cohort (mean age, 64 years; 51% female) consisting of 500 consecutive patients who underwent routine chest CT served as the training set for a DL model that was used to measure left atrial volume. The model was then used to measure atrial size for a separate 500-patient validation cohort (mean age, 61 years; 46% female), in which the AF status was determined by performing a chart review. The performance of automated atrial size as a predictor of AF was evaluated by using a receiver operating characteristic analysis.
There was good agreement between manual and model-generated segmentation maps by all measures of overlap and surface distance (mean Dice = 0.87, intersection over union = 0.77, Hausdorff distance = 4.36 mm, average symmetric surface distance = 0.96 mm), and agreement was slightly but significantly greater than that between human observers (mean Dice = 0.85 [automated] vs 0.84 [manual]; = .004). Atrial volume was a good predictor of AF in the validation cohort (area under the receiver operating characteristic curve = 0.768) and was an independent predictor of AF, with an age-adjusted relative risk of 2.9.
Left atrial volume is an independent predictor of the AF status as measured at routine nongated chest CT. Deep learning is a suitable tool for automated measurement.© RSNA, 2019See also the commentary by de Roos and Tao in this issue.
测试深度学习（DL）模型在常规非门控胸部CT中预测心房颤动（AF）的性能。
一个回顾性推导队列（平均年龄64岁；51%为女性），由500例连续接受常规胸部CT检查的患者组成，作为用于测量左心房容积的DL模型的训练集。然后使用该模型测量另一个由500例患者组成的验证队列（平均年龄61岁；46%为女性）的心房大小，通过查阅病历确定该队列中的AF状态。通过使用受试者操作特征分析评估自动测量的心房大小作为AF预测指标的性能。
通过所有重叠和表面距离测量方法，手动分割图与模型生成的分割图之间具有良好的一致性（平均Dice系数 = 0.87，交并比 = 0.77，豪斯多夫距离 = 4.36 mm，平均对称表面距离 = 0.96 mm），且一致性略高于但显著高于人类观察者之间的一致性（平均Dice系数 = 0.85 [自动] 对0.84 [手动]；P = 0.004）。在验证队列中，心房容积是AF的良好预测指标（受试者操作特征曲线下面积 = 0.768），并且是AF的独立预测指标，年龄调整后的相对风险为2.9。
在常规非门控胸部CT测量中，左心房容积是AF状态的独立预测指标。深度学习是自动测量的合适工具。© RSNA，2019另见本期德罗os和陶的评论。
