An externally validated fully automated deep learning algorithm to classify COVID-19 and other pneumonias on chest computed tomography
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...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
European Respiratory Society
2022-05-01
|
| Series: | ERJ Open Research |
| Online Access: | http://openres.ersjournals.com/content/8/2/00579-2021.full |
| _version_ | 1797809410113798144 |
|---|---|
| author | Akshayaa Vaidyanathan Julien Guiot Fadila Zerka Flore Belmans Ingrid Van Peufflik Louis Deprez Denis Danthine Gregory Canivet Philippe Lambin Sean Walsh Mariaelena Occhipinti Paul Meunier Wim Vos Pierre Lovinfosse Ralph T.H. Leijenaar |
| author_facet | Akshayaa Vaidyanathan Julien Guiot Fadila Zerka Flore Belmans Ingrid Van Peufflik Louis Deprez Denis Danthine Gregory Canivet Philippe Lambin Sean Walsh Mariaelena Occhipinti Paul Meunier Wim Vos Pierre Lovinfosse Ralph T.H. Leijenaar |
| author_sort | Akshayaa Vaidyanathan |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-03-13T06:52:17Z |
| format | Article |
| id | doaj.art-06c0f2983a8e45f79819d1d0caf8e1e8 |
| institution | Directory Open Access Journal |
| issn | 2312-0541 |
| language | English |
| last_indexed | 2024-03-13T06:52:17Z |
| publishDate | 2022-05-01 |
| publisher | European Respiratory Society |
| record_format | Article |
| series | ERJ Open Research |
| spelling | doaj.art-06c0f2983a8e45f79819d1d0caf8e1e82023-06-07T13:30:08ZengEuropean Respiratory SocietyERJ Open Research2312-05412022-05-018210.1183/23120541.00579-202100579-2021An externally validated fully automated deep learning algorithm to classify COVID-19 and other pneumonias on chest computed tomographyAkshayaa Vaidyanathan0Julien Guiot1Fadila Zerka2Flore Belmans3Ingrid Van Peufflik4Louis Deprez5Denis Danthine6Gregory Canivet7Philippe Lambin8Sean Walsh9Mariaelena Occhipinti10Paul Meunier11Wim Vos12Pierre Lovinfosse13Ralph T.H. Leijenaar14 Radiomics (Oncoradiomics SA), Liège, Belgium Dept of Pneumology, University Hospital of Liège, Liège, Belgium Radiomics (Oncoradiomics SA), Liège, Belgium Radiomics (Oncoradiomics SA), Liège, Belgium Radiomics (Oncoradiomics SA), Liège, Belgium Dept of Radiology, University Hospital of Liège, Liège, Belgium Dept of Radiology, University Hospital of Liège, Liège, Belgium Dept of Computer Applications, University Hospital of Liège, Liège, Belgium The D-Lab, Depts of Precision Medicine, and Radiology and Nuclear Medicine, GROW–School for Oncology, Maastricht University, Maastricht, The Netherlands Radiomics (Oncoradiomics SA), Liège, Belgium Radiomics (Oncoradiomics SA), Liège, Belgium Dept of Radiology, University Hospital of Liège, Liège, Belgium Radiomics (Oncoradiomics SA), Liège, Belgium Dept of Nuclear Medicine and Oncological Imaging, University Hospital of Liège, Liège, Belgium Radiomics (Oncoradiomics SA), Liège, Belgium 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.http://openres.ersjournals.com/content/8/2/00579-2021.full |
| spellingShingle | Akshayaa Vaidyanathan Julien Guiot Fadila Zerka Flore Belmans Ingrid Van Peufflik Louis Deprez Denis Danthine Gregory Canivet Philippe Lambin Sean Walsh Mariaelena Occhipinti Paul Meunier Wim Vos Pierre Lovinfosse Ralph T.H. Leijenaar An externally validated fully automated deep learning algorithm to classify COVID-19 and other pneumonias on chest computed tomography ERJ Open Research |
| title | An externally validated fully automated deep learning algorithm to classify COVID-19 and other pneumonias on chest computed tomography |
| title_full | An externally validated fully automated deep learning algorithm to classify COVID-19 and other pneumonias on chest computed tomography |
| title_fullStr | An externally validated fully automated deep learning algorithm to classify COVID-19 and other pneumonias on chest computed tomography |
| title_full_unstemmed | An externally validated fully automated deep learning algorithm to classify COVID-19 and other pneumonias on chest computed tomography |
| title_short | An externally validated fully automated deep learning algorithm to classify COVID-19 and other pneumonias on chest computed tomography |
| title_sort | externally validated fully automated deep learning algorithm to classify covid 19 and other pneumonias on chest computed tomography |
| url | http://openres.ersjournals.com/content/8/2/00579-2021.full |
| work_keys_str_mv | AT akshayaavaidyanathan anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT julienguiot anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT fadilazerka anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT florebelmans anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT ingridvanpeufflik anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT louisdeprez anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT denisdanthine anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT gregorycanivet anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT philippelambin anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT seanwalsh anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT mariaelenaocchipinti anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT paulmeunier anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT wimvos anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT pierrelovinfosse anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT ralphthleijenaar anexternallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT akshayaavaidyanathan externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT julienguiot externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT fadilazerka externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT florebelmans externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT ingridvanpeufflik externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT louisdeprez externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT denisdanthine externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT gregorycanivet externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT philippelambin externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT seanwalsh externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT mariaelenaocchipinti externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT paulmeunier externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT wimvos externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT pierrelovinfosse externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography AT ralphthleijenaar externallyvalidatedfullyautomateddeeplearningalgorithmtoclassifycovid19andotherpneumoniasonchestcomputedtomography |