深度新冠检测：一项使用胸部CT进行新冠病毒肺炎肺部检测与预后评估的国际经验。

Deep COVID DeteCT: an international experience on COVID-19 lung detection and prognosis using chest CT.

作者信息

Lee Edward H, Zheng Jimmy, Colak Errol, Mohammadzadeh Maryam, Houshmand Golnaz, Bevins Nicholas, Kitamura Felipe, Altinmakas Emre, Reis Eduardo Pontes, Kim Jae-Kwang, Klochko Chad, Han Michelle, Moradian Sadegh, Mohammadzadeh Ali, Sharifian Hashem, Hashemi Hassan, Firouznia Kavous, Ghanaati Hossien, Gity Masoumeh, Doğan Hakan, Salehinejad Hojjat, Alves Henrique, Seekins Jayne, Abdala Nitamar, Atasoy Çetin, Pouraliakbar Hamidreza, Maleki Majid, Wong S Simon, Yeom Kristen W

机构信息

Department of Radiology, School of Medicine, Stanford University, Stanford, CA, 94305, USA.

Unity Health Toronto, University of Toronto, Toronto, ON, M5S, Canada.

出版信息

NPJ Digit Med. 2021 Jan 29;4(1):11. doi: 10.1038/s41746-020-00369-1.

The Coronavirus disease 2019 (COVID-19) presents open questions in how we clinically diagnose and assess disease course. Recently, chest computed tomography (CT) has shown utility for COVID-19 diagnosis. In this study, we developed Deep COVID DeteCT (DCD), a deep learning convolutional neural network (CNN) that uses the entire chest CT volume to automatically predict COVID-19 (COVID+) from non-COVID-19 (COVID-) pneumonia and normal controls. We discuss training strategies and differences in performance across 13 international institutions and 8 countries. The inclusion of non-China sites in training significantly improved classification performance with area under the curve (AUCs) and accuracies above 0.8 on most test sites. Furthermore, using available follow-up scans, we investigate methods to track patient disease course and predict prognosis.

2019冠状病毒病（COVID-19）在我们如何进行临床诊断和评估疾病进程方面存在一些悬而未决的问题。最近，胸部计算机断层扫描（CT）已显示出对COVID-19诊断的效用。在本研究中，我们开发了深度COVID检测（DCD），这是一种深度学习卷积神经网络（CNN），它使用整个胸部CT容积来自动从非COVID-19（COVID-）肺炎和正常对照中预测COVID-19（COVID+）。我们讨论了13个国际机构和8个国家的训练策略和性能差异。在训练中纳入非中国地区的样本显著提高了分类性能，在大多数测试地点的曲线下面积（AUC）和准确率均高于0.8。此外，利用现有的随访扫描，我们研究了跟踪患者疾病进程和预测预后的方法。