AI-driven quantification, staging and outcome prediction of COVID-19 pneumonia

被引：98

作者：

Chassagnon, Guillaume ^{[1
,2
,3
]}

Vakalopoulou, Maria ^{[4
,5
,6
,7
]}

Battistella, Enzo ^{[4
,5
,6
,7
,8
]}

Christodoulidis, Stergios ^{[9
,10
]}

Trieu-Nghi Hoang-Thi ^{[1
]}

Dangeard, Severine ^{[1
]}

Deutsch, Eric ^{[7
,8
]}

Andre, Fabrice ^{[9
,10
]}

Guillo, Enora ^{[1
]}

Halm, Nara ^{[1
]}

El Hajj, Stefany ^{[1
]}

Bompard, Florian ^{[1
]}

Neveu, Sophie ^{[1
]}

Hani, Chahinez ^{[1
]}

Saab, Ines ^{[1
]}

Campredon, Alienor ^{[1
]}

Koulakian, Hasmik ^{[1
]}

Bennani, Souhail ^{[1
]}

Freche, Gael ^{[1
]}

Barat, Maxime ^{[1
,2
]}

Lombard, Aurelien ^{[11
]}

Fournier, Laure ^{[2
,12
]}

Monnier, Hippolyte ^{[12
]}

Grand, Teodor ^{[12
]}

Gregory, Jules ^{[2
,13
]}

Nguyen, Yann ^{[2
,14
]}

Khalil, Antoine ^{[2
,15
]}

Mahdjoub, Elyas ^{[2
,15
]}

Brillet, Pierre-Yves ^{[16
,17
]}

Ba, Stephane Tran ^{[16
,17
]}

Bousson, Valerie ^{[2
,18
]}

Mekki, Ahmed ^{[19
,20
,21
]}

Carlier, Robert-Yves ^{[19
,20
,21
]}

Revel, Marie-Pierre ^{[1
,2
,3
]}

Paragios, Nikos ^{[4
,5
,7
,11
]}

机构：

[1] Hop Cochin, AP HP, Ctr Univ Paris, Radiol Dept, 27 Rue Faubourg St Jacques, F-75014 Paris, France

[2] Univ Paris, 85 Blvd St Germain, F-75006 Paris, France

[3] Inst Cochin, INSERM, U1016, 22 Rue Mechain, F-75014 Paris, France

[4] Univ Paris Saclay, Cent Supelec, Math & Informat Complexite & Syst, Gif Sur Yvette, France

[5] 3 Rue Joliot Curie, F-91190 Gif Sur Yvette, France

[6] Inria Saclay, F-91190 Gif Sur Yvette, France

[7] Gustave Roussy Cent Supelec TheraPanacea, Noesia Ctr Artificial Intelligence Radiat Therapy, Gustave Roussy Canc Campus, Villejuif, France

[8] Univ Paris Saclay, Inst Gustave Roussy, INSERM, Mol Radiotherapy & Innovat Therapeut U1030, 114 Rue Edouard Vaillant, F-94800 Villejuif, France

[9] Univ Paris Saclay, Inst Gustave Roussy, INSERM, Predict Biomarkers & New Therapeut Strategies Onc, 114 Rue Edouard Vaillant, F-94800 Villejuif, France

[10] Univ Paris Saclay, Inst Gustave Roussy, Prism Precis Med Ctr, 114 Rue Edouard Vaillant, F-94800 Villejuif, France

[11] TheraPanacea, 29 Rue Faubourg St Jacques, F-75014 Paris, France

[12] Hop Europeen Georges Pompidou, AP HP, Ctr Univ Paris, Radiol Dept, 20 Rue Univ Paris Saclay, F-75015 Paris, France

[13] Nord Univ Paris, Hop Beaujon, AP HP, Radiol Dept, 100 Blvd Gen Leclerc, F-92110 Clichy, France

[14] Nord Univ Paris, Hop Beaujon, AP HP, Internal Med Dept, 100 Blvd Gen Leclerc, F-92110 Clichy, France

[15] Nord Univ Paris, Hop Bichat, AP HP, Radiol Dept, 46 Rue Henri Huchard, F-75018 Paris, France

[16] Hop Univ Paris Seine St Denis, Hop Avicenne, AP HP, Radiol Dept, 125 Rue Stalingrad, F-93000 Bobigny, France

[17] Univ Sorbonne Paris Nord, 99 Ave Jean Baptiste Clement, F-93430 Villetaneuse, France

[18] Nord Univ Paris, Hop Lariboisiere, AP HP, Radiol Dept, 2 Rue Ambroise Pare, F-75010 Paris, France

[19] Univ Paris Saclay, Hop Ambroise Pare, AP HP, Radiol Dept, 9 Ave Charles de Gaulle, F-92100 Boulogne, France

[20] Univ Paris Saclay, Raymond Pointcare AP HP, Radiol Dept, 104 Blvd Raymond Poincare, F-92380 Garches, France

[21] Univ Paris Saclay, Espace Technol Bat Discovery RD 128 2e &, F-91190 St Aubin, France

来源：

MEDICAL IMAGE ANALYSIS | 2021年 / 67卷

基金：

瑞士国家科学基金会;

关键词：

COVID; 19; pneumonia; Artifial intelligence; Deep learning; Staging; Prognosis; Biomarker discovery; Ensemble methods; CLASSIFICATION;

D O I：

10.1016/j.media.2020.101860

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Coronavirus disease 2019 (COVID-19) emerged in 2019 and disseminated around the world rapidly. Computed tomography (CT) imaging has been proven to be an important tool for screening, disease quantification and staging. The latter is of extreme importance for organizational anticipation (availability of intensive care unit beds, patient management planning) as well as to accelerate drug development through rapid, reproducible and quantified assessment of treatment response. Even if currently there are no specific guidelines for the staging of the patients, CT together with some clinical and biological biomarkers are used. In this study, we collected a multi-center cohort and we investigated the use of medical imaging and artificial intelligence for disease quantification, staging and outcome prediction. Our approach relies on automatic deep learning-based disease quantification using an ensemble of architectures, and a data driven consensus for the staging and outcome prediction of the patients fusing imaging biomarkers with clinical and biological attributes. Highly promising results on multiple external/independent evaluation cohorts as well as comparisons with expert human readers demonstrate the potentials of our approach. (C) 2020 The Author(s). Published by Elsevier B.V.

引用

页数：16

共 44 条

[1]

Abadi Martin, 2016, arXiv

[2] Semantic Segmenation of Pathological Lung Tissue With Dilated Fully Convolutional Networks [J].