摘要BACKGROUND:Coronavirus Disease 2019(COVID-19)is currently a global pandemic.Information about predicting mortality in severe COVID-19 remains unclear.METHODS:A total of 151 COVID-19 in-patients from January 23 to March 8,2020,were divided into severe and critically severe groups and survival and mortality groups.Differences in the clinical and imaging data between the groups were analyzed.Factors associated with COVID-19 mortality were analyzed by logistic regression,and a mortality prediction model was developed.RESULTS:Many clinical and imaging indices were significantly different between groups,including age,epidemic history,medical history,duration of symptoms before admission,routine blood parameters,inflammatory-related factors,Na+,myocardial zymogram,liver and renal function,coagulation function,fraction of inspired oxygen and complications.The proportions of patients with imaging Stage III and a comprehensive computed tomography score were significantly increased in the mortality group.Factors in the prediction model included patient age,cardiac injury,acute kidney injury,and acute respiratory distress syndrome.The area under the receiver operating characteristic curve of the prediction model was 0.9593.CONCLUSIONS:The clinical and imaging data reflected the severity of COVID-19 pneumonia.The mortality prediction model might be a promising method to help clinicians quickly identify COVID-19 patients who are at high risk of death.