TY - JOUR
T1 - Development of a clinical decision support system for severity risk prediction and triage of COVID-19 patients at hospital admission
T2 - an international multicentre study
AU - Wu, Guangyao
AU - Yang, Pei
AU - Xie, Yuanliang
AU - Woodruff, Henry C.
AU - Rao, Xiangang
AU - Guiot, Julien
AU - Frix, Anne-Noelle
AU - Louis, Renaud
AU - Moutschen, Michel
AU - Li, Jiawei
AU - Li, Jing
AU - Yan, Chenggong
AU - Du, Dan
AU - Zhao, Shengchao
AU - Ding, Yi
AU - Liu, Bin
AU - Sun, Wenwu
AU - Albarello, Fabrizio
AU - D'Abramo, Alessandra
AU - Schinina, Vincenzo
AU - Nicastri, Emanuele
AU - Occhipinti, Mariaelena
AU - Barisione, Giovanni
AU - Barisione, Emanuela
AU - Halilaj, Iva
AU - Lovinfosse, Pierre
AU - Wang, Xiang
AU - Wu, Jianlin
AU - Lambin, Philippe
N1 - Funding Information:
Conflict of interest: G. Wu has nothing to disclose. P. Yang has nothing to disclose. Y. Xie has nothing to disclose. H.C. Woodruff owns shares in Oncoradiomics, outside the submitted work. X. Rao has nothing to disclose. J. Guiot has nothing to disclose. A-N. Frix has nothing to disclose. R. Louis has nothing to disclose. M. Moutschen has nothing to disclose. Jiawei Li has nothing to disclose. Jing Li has nothing to disclose. C. Yan has nothing to disclose. D. Du has nothing to disclose. S. Zhao has nothing to disclose. Y. Ding has nothing to disclose. B. Liu has nothing to disclose. W. Sun has nothing to disclose. F. Albarello has nothing to disclose. A. D’Abramo has nothing to disclose. V. Schininà has nothing to disclose. E. Nicastri has nothing to disclose. M. Occhipinti reports grants from Menarini Foundation and Novartis, outside the submitted work. G. Barisione has nothing to disclose. E. Barisione has nothing to disclose. I. Halilaj has nothing to disclose. P. Lovinfosse has nothing to disclose. X. Wang has nothing to disclose. J. Wu has nothing to disclose. P. Lambin reports, within the submitted work, minority shares in The Medical Cloud Company and, outside the submitted work, grants/sponsored research agreements from Varian medical, Oncoradiomics, ptTheragnostic/DNAmito and, Health Innovation Ventures; he received an advisor/presenter fee and/or reimbursement of travel costs/external grant writing fee and/or in kind manpower contribution from Oncoradiomics, BHV, Varian, Elekta, ptTheragnostic and Convert pharmaceuticals; P. Lambin has shares in the company Oncoradiomics SA, Convert pharmaceuticals SA and is co-inventor of two issued patents with royalties on radiomics (PCT/NL2014/050248, PCT/ NL2014/050728) licensed to Oncoradiomics and one issued patent on mtDNA (PCT/EP2014/059089) licensed to ptTheragnostic/DNAmito, three non-patented invention (softwares) licensed to ptTheragnostic/DNAmito, Oncoradiomics and Health Innovation Ventures and three non-issues, non licensed patents on Deep Learning-Radiomics and LSRT (N2024482, N2024889, N2024889).
Funding Information:
Support statement: This work was supported from ERC advanced grant (ERC-ADG-2015, number 694812 – Hypoximmuno), European Program H2020 (ImmunoSABR – number 733008, PREDICT – ITN – number 766276, CHAIMELEON – number 952172, EuCanImage – number 952103), TRANSCAN Joint Transnational Call 2016 ( JTC2016 “CLEARLY” – number UM 2017-8295), China Scholarships Council (number 201808210318), and Interreg V-A Euregio Meuse-Rhine (“Euradiomics” – number EMR4). The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. This work was supported by the Dutch Cancer Society (KWF Kankerbestrijding), Project number 12085/2018-2. Funding information for this article has been deposited with the Crossref Funder Registry.
Publisher Copyright:
Copyright © ERS 2020. This version is distributed under the terms of the Creative Commons Attribution NonCommercial Licence 4.0.
PY - 2020/8/1
Y1 - 2020/8/1
N2 - Background: The outbreak of coronavirus disease 2019 (COVID-19) has globally strained medical resources and caused significant mortality.Objective To develop and validate a machine-learning model based on clinical features for severity risk assessment and triage for COVID-19 patients at hospital admission.Method: 725 patients were used to train and validate the model. This included a retrospective cohort from Wuhan, China of 299 hospitalised COVID-19 patients from 23 December 2019 to 13 February 2020, and five cohorts with 426 patients from eight centres in China, Italy and Belgium from 20 February 2020 to 21 March 2020. The main outcome was the onset of severe or critical illness during hospitalisation. Model performances were quantified using the area under the receiver operating characteristic curve (AUC) and metrics derived from the confusion matrix.Results: In the retrospective cohort, the median age was 50 years and 137 (45.8%) were male. In the five test cohorts, the median age was 62 years and 236 (55.4%) were male. The model was prospectively validated on five cohorts yielding AUCs ranging from 0.84 to 0.93, with accuracies ranging from 74.4% to 87.5%, sensitivities ranging from 75.0% to 96.9%, and specificities ranging from 55.0% to 88.0%, most of which performed better than the pneumonia severity index. The cut-off values of the low-, medium- and high-risk probabilities were 0.21 and 0.80. The online calculators can be found at www.covid19risk.ai.Conclusion: The machine-learning model, nomogram and online calculator might be useful to access the onset of severe and critical illness among COVID-19 patients and triage at hospital admission.
AB - Background: The outbreak of coronavirus disease 2019 (COVID-19) has globally strained medical resources and caused significant mortality.Objective To develop and validate a machine-learning model based on clinical features for severity risk assessment and triage for COVID-19 patients at hospital admission.Method: 725 patients were used to train and validate the model. This included a retrospective cohort from Wuhan, China of 299 hospitalised COVID-19 patients from 23 December 2019 to 13 February 2020, and five cohorts with 426 patients from eight centres in China, Italy and Belgium from 20 February 2020 to 21 March 2020. The main outcome was the onset of severe or critical illness during hospitalisation. Model performances were quantified using the area under the receiver operating characteristic curve (AUC) and metrics derived from the confusion matrix.Results: In the retrospective cohort, the median age was 50 years and 137 (45.8%) were male. In the five test cohorts, the median age was 62 years and 236 (55.4%) were male. The model was prospectively validated on five cohorts yielding AUCs ranging from 0.84 to 0.93, with accuracies ranging from 74.4% to 87.5%, sensitivities ranging from 75.0% to 96.9%, and specificities ranging from 55.0% to 88.0%, most of which performed better than the pneumonia severity index. The cut-off values of the low-, medium- and high-risk probabilities were 0.21 and 0.80. The online calculators can be found at www.covid19risk.ai.Conclusion: The machine-learning model, nomogram and online calculator might be useful to access the onset of severe and critical illness among COVID-19 patients and triage at hospital admission.
KW - ADULTS
KW - CORONAVIRUS
KW - PNEUMONIA
KW - WUHAN
U2 - 10.1183/13993003.01104-2020
DO - 10.1183/13993003.01104-2020
M3 - Article
C2 - 32616597
SN - 0903-1936
VL - 56
JO - European Respiratory Journal
JF - European Respiratory Journal
IS - 2
M1 - 2001104
ER -