No abstract

with a 1-day history of fever without dizziness, cough, and headaches. On presentation, his temperature was 38•1°C. Laboratory tests showed a C-reactive protein concentration of 0•56 mg/dL (normal range 0•00-0•60] mg/dL). Complete blood count showed elevated leukocytes (10 060 cells per μL [normal range 3500-9500 cells per μL]), neutrophils (7550 cells per μL [1800-6300 cells per μL]), and monocytes (990 cells per μL [100-600 cells per μL]), while the lymphocyte count (1490 cells per μL) was in the normal range (1100-3200 cells per μL). The patient was negative for influenza A and B viruses, adenovirus, respiratory syncytial virus, and parainfluenza 1, 2, and 3 viruses. Chest CT showed multiple ground-glass opacities in the lower lobes bilaterally.The patient was given antibacterial, antiviral, and corticosteroid treatments (moxifloxacin [0•4 g/day] for 5 days, followed by ribavirin [0•5 g/day] and methylprednisolone [40 mg/day] for 5 days) via intravenous drop infusion. However, after 10 days, the patient had persistent fever (highest temperature 38•5°C), cough, and shortness of breath. The patient was diagnosed with coronavirus Contributors CZ and CG contributed to data analysis, data interpretation, the literature search, and manuscript drafting. YX contributed to data collection, data analysis, and figure preparation. MX contributed to study design and reviewed the final draft. All authors read and approved the manuscript.

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.