Deep COVID DeteCT: an international experience on COVID-19 lung detection and prognosis using chest CT
Edward H. Lee, Jimmy Zheng, Errol Colak, Maryam Mohammadzadeh, Golnaz Houshmand, Nicholas Bevins, Felipe Kitamura, Emre Altınmakas, Eduardo Pontes Reis, Jae-Kwang Kim, Chad Klochko, Michelle Han, Sadegh Moradian, Ali Mohammadzadeh, Hashem Sharifian, Hassan Hashemi, Kavous Firouznia, Hossien Ghanaati, Masoumeh Gity, Hakan Doğan, Hojjat Salehinejad, Henrique Alves, Jayne Seekins, Nitamar Abdala, Çetin Atasoy, Hamidreza Pouraliakbar, Majid Maleki, S.S. Wong, Kristen W. Yeom
Abstract
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.