tetano
Editor, Senior Moderator
NPJ Digit Med
. 2021 Jan 29;4(1):11.
doi: 10.1038/s41746-020-00369-1.
Deep COVID DeteCT: an international experience on COVID-19 lung detection and prognosis using chest CT
Edward H Lee[SUP] 1 [/SUP], Jimmy Zheng[SUP] 2 [/SUP], Errol Colak[SUP] 3 [/SUP], Maryam Mohammadzadeh[SUP] 4 [/SUP], Golnaz Houshmand[SUP] 5 [/SUP], Nicholas Bevins[SUP] 6 [/SUP], Felipe Kitamura[SUP] 7 [/SUP], Emre Altinmakas[SUP] 8 [/SUP], Eduardo Pontes Reis[SUP] 9 [/SUP], Jae-Kwang Kim[SUP] 10 [/SUP], Chad Klochko[SUP] 5 [/SUP], Michelle Han[SUP] 2 [/SUP], Sadegh Moradian[SUP] 11 [/SUP], Ali Mohammadzadeh[SUP] 5 [/SUP], Hashem Sharifian[SUP] 4 [/SUP], Hassan Hashemi[SUP] 12 [/SUP], Kavous Firouznia[SUP] 12 [/SUP], Hossien Ghanaati[SUP] 12 [/SUP], Masoumeh Gity[SUP] 12 [/SUP], Hakan Doğan[SUP] 8 [/SUP], Hojjat Salehinejad[SUP] 3 [/SUP], Henrique Alves[SUP] 7 [/SUP], Jayne Seekins[SUP] 2 [/SUP], Nitamar Abdala[SUP] 7 [/SUP], ?etin Atasoy[SUP] 8 [/SUP], Hamidreza Pouraliakbar[SUP] 5 [/SUP], Majid Maleki[SUP] 5 [/SUP], S Simon Wong[SUP] 13 [/SUP], Kristen W Yeom[SUP] 14 [/SUP]
Affiliations
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.
. 2021 Jan 29;4(1):11.
doi: 10.1038/s41746-020-00369-1.
Deep COVID DeteCT: an international experience on COVID-19 lung detection and prognosis using chest CT
Edward H Lee[SUP] 1 [/SUP], Jimmy Zheng[SUP] 2 [/SUP], Errol Colak[SUP] 3 [/SUP], Maryam Mohammadzadeh[SUP] 4 [/SUP], Golnaz Houshmand[SUP] 5 [/SUP], Nicholas Bevins[SUP] 6 [/SUP], Felipe Kitamura[SUP] 7 [/SUP], Emre Altinmakas[SUP] 8 [/SUP], Eduardo Pontes Reis[SUP] 9 [/SUP], Jae-Kwang Kim[SUP] 10 [/SUP], Chad Klochko[SUP] 5 [/SUP], Michelle Han[SUP] 2 [/SUP], Sadegh Moradian[SUP] 11 [/SUP], Ali Mohammadzadeh[SUP] 5 [/SUP], Hashem Sharifian[SUP] 4 [/SUP], Hassan Hashemi[SUP] 12 [/SUP], Kavous Firouznia[SUP] 12 [/SUP], Hossien Ghanaati[SUP] 12 [/SUP], Masoumeh Gity[SUP] 12 [/SUP], Hakan Doğan[SUP] 8 [/SUP], Hojjat Salehinejad[SUP] 3 [/SUP], Henrique Alves[SUP] 7 [/SUP], Jayne Seekins[SUP] 2 [/SUP], Nitamar Abdala[SUP] 7 [/SUP], ?etin Atasoy[SUP] 8 [/SUP], Hamidreza Pouraliakbar[SUP] 5 [/SUP], Majid Maleki[SUP] 5 [/SUP], S Simon Wong[SUP] 13 [/SUP], Kristen W Yeom[SUP] 14 [/SUP]
Affiliations
- PMID: 33514852
- DOI: 10.1038/s41746-020-00369-1
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.