Article (Scientific journals)
An externally validated fully automated deep learning algorithm to classify COVID-19 and other pneumonias on chest computed tomography.
Vaidyanathan, Akshayaa; Guiot, Julien; Zerka, Fadila et al.
2022In ERJ Open Research, 8 (2), p. 00579-2021
Peer Reviewed verified by ORBi
 

Files


Full Text
00579-2021.full.pdf
Author postprint (813.89 kB)
Download

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Pulmonary and Respiratory Medicine
Abstract :
[en] Purpose: In this study, we propose an artificial intelligence (AI) framework based on three-dimensional convolutional neural networks to classify computed tomography (CT) scans of patients with coronavirus disease 2019 (COVID-19), influenza/community-acquired pneumonia (CAP), and no infection, after automatic segmentation of the lungs and lung abnormalities. Methods: The AI classification model is based on inflated three-dimensional Inception architecture and was trained and validated on retrospective data of CT images of 667 adult patients (no infection n=188, COVID-19 n=230, influenza/CAP n=249) and 210 adult patients (no infection n=70, COVID-19 n=70, influenza/CAP n=70), respectively. The model's performance was independently evaluated on an internal test set of 273 adult patients (no infection n=55, COVID-19 n= 94, influenza/CAP n=124) and an external validation set from a different centre (305 adult patients: COVID-19 n=169, no infection n=76, influenza/CAP n=60). Results: The model showed excellent performance in the external validation set with area under the curve of 0.90, 0.92 and 0.92 for COVID-19, influenza/CAP and no infection, respectively. The selection of the input slices based on automatic segmentation of the abnormalities in the lung reduces analysis time (56 s per scan) and computational burden of the model. The Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) score of the proposed model is 47% (15 out of 32 TRIPOD items). Conclusion: This AI solution provides rapid and accurate diagnosis in patients suspected of COVID-19 infection and influenza.
Disciplines :
Radiology, nuclear medicine & imaging
Author, co-author :
Vaidyanathan, Akshayaa ;  Radiomics (Oncoradiomics SA), Liège, Belgium ; The D-Lab, Depts of Precision Medicine, and Radiology and Nuclear Medicine, GROW-School for Oncology, Maastricht University, Maastricht, The Netherlands ; These authors have contributed equally to this work and share first authorship
Guiot, Julien  ;  Centre Hospitalier Universitaire de Liège - CHU > > Service de pneumologie - allergologie ; These authors have contributed equally to this work and share first authorship
Zerka, Fadila;  Radiomics (Oncoradiomics SA), Liège, Belgium ; The D-Lab, Depts of Precision Medicine, and Radiology and Nuclear Medicine, GROW-School for Oncology, Maastricht University, Maastricht, The Netherlands
Belmans, Flore;  Radiomics (Oncoradiomics SA), Liège, Belgium
Van Peufflik, Ingrid;  Radiomics (Oncoradiomics SA), Liège, Belgium
Deprez, Louis ;  Centre Hospitalier Universitaire de Liège - CHU > > Service médical de radiodiagnostic
Danthine, Denis  ;  Centre Hospitalier Universitaire de Liège - CHU > > Service médical de médecine nucléaire et imagerie onco
CANIVET, Grégory ;  Centre Hospitalier Universitaire de Liège - CHU > > Service des Applications Informatiques (SAI)
Lambin, Philippe ;  The D-Lab, Depts of Precision Medicine, and Radiology and Nuclear Medicine, GROW-School for Oncology, Maastricht University, Maastricht, The Netherlands
Walsh, Sean;  Radiomics (Oncoradiomics SA), Liège, Belgium
Occhipinti, Mariaelena;  Radiomics (Oncoradiomics SA), Liège, Belgium
Meunier, Paul ;  Centre Hospitalier Universitaire de Liège - CHU > > Service médical de radiodiagnostic
Vos, Wim;  Radiomics (Oncoradiomics SA), Liège, Belgium
Lovinfosse, Pierre ;  Centre Hospitalier Universitaire de Liège - CHU > > Service médical de médecine nucléaire et imagerie onco
Leijenaar, Ralph T H ;  Radiomics (Oncoradiomics SA), Liège, Belgium
More authors (5 more) Less
Language :
English
Title :
An externally validated fully automated deep learning algorithm to classify COVID-19 and other pneumonias on chest computed tomography.
Publication date :
April 2022
Journal title :
ERJ Open Research
eISSN :
2312-0541
Publisher :
European Respiratory Society, England
Volume :
8
Issue :
2
Pages :
00579-2021
Peer reviewed :
Peer Reviewed verified by ORBi
European Projects :
H2020 - 101005122 - DRAGON - The RapiD and SecuRe AI enhAnced DiaGnosis, Precision Medicine and Patient EmpOwerment Centered Decision Support System for Coronavirus PaNdemics
Funders :
EU - European Union [BE]
Funding number :
European Marie Curie grant (PREDICT – ITN, number 766276)
Funding text :
The authors thank Fabio Bottari (Radiomics, Liège, Belgium) for providing medical writing support in accordance with Good Publication Practice (GPP3) guidelines. Support statement: The authors acknowledge financial support from European Marie Curie grant (PREDICT – ITN, number 766276) and the European Union’s Horizon 2020 research and innovation programme under grant agreements DRAGON – 101005122 (call: H2020-JTI-IMI2-2020-21) and iCOVID – 101016131 (call: H2020-SC1-PHE-CORONAVIRUS-2020-2) for the execution of this work. Funding information for this article has been deposited with the Crossref Funder Registry.Support statement: The authors acknowledge financial support from European Marie Curie grant (PREDICT – ITN, number 766276) and the European Union’s Horizon 2020 research and innovation programme under grant agreements DRAGON – 101005122 (call: H2020-JTI-IMI2-2020-21) and iCOVID – 101016131 (call: H2020-SC1-PHE-CORONAVIRUS-2020-2) for the execution of this work. Funding information for this article has been deposited with the Crossref Funder Registry.
Available on ORBi :
since 31 August 2022

Statistics


Number of views
75 (13 by ULiège)
Number of downloads
23 (3 by ULiège)

Scopus citations®
 
6
Scopus citations®
without self-citations
5
OpenCitations
 
0

Bibliography


Similar publications



Contact ORBi