BackgroundThe outbreak of the coronavirus disease 2019 (COVID-19) has globally strained medical resources and caused significant mortality.ObjectiveTo develop and validate machine-learning model based on clinical features for severity risk assessment and triage for COVID-19 patients at hospital admission.Method725 patients were used to train and validate the model including a retrospective cohort of 299 hospitalised COVID-19 patients at Wuhan, China, from December 23, 2019, to February 13, 2020, and five cohorts with 426 patients from eight centers in China, Italy, and Belgium, from February 20, 2020, to March 21, 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.ResultsThe median age was 50.0 years and 137 (45.8%) were men in the retrospective cohort. The median age was 62.0 years and 236 (55.4%) were men in five cohorts. The model was prospectively validated on five cohorts yielding AUCs ranging from 0.84 to 0.89, with accuracies ranging from 74.4% to 87.5%, sensitivities ranging from 75.0% to 96.9%, and specificities ranging from 57.5% to 88.0%, all 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.ConclusionThe 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.

Word count: 2973All rights reserved. No reuse allowed without permission.was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Key points:Question How do nomograms and machine-learning algorithms of severity risk prediction and triage of COVID- patients at hospital admission perform?Findings This model was prospectively validated on six test datasets comprising of 426 patients and yielded AUCs ranging from 0.816 to 0.976, accuracies ranging from 70.8% to 93.8%, sensitivities ranging from 83.7% to 100%, and specificities ranging from 41.0% to 95.7%. The cut-off probability values for low, medium, and high-risk groups were 0.072 and 0.244.Meaning The findings of this study suggest that our models performs well for the diagnosis and prediction of progression to severe or critical illness of COVID-19 patients and could be used for triage of COVID-19 patients at hospital admission.All rights reserved. No reuse allowed without permission.was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

No abstract