LAUSR

21 To develop a machine learning model and nomogram to predict the probability of persistent virus 22 shedding (PVS) in hospitalized patients with coronavirus disease 2019 (COVID-19), the clinical 23 symptoms and signs, laboratory parameters, cytokines, and immune cell data of 429 patients with 24 nonsevere COVID-19 were retrospectively reviewed. Two models were developed using the 25 Akaike information criterion (AIC). The performance of these two models was analyzed and 26 compared by the receiver operating characteristic (ROC) curve, calibration curve, net 27 reclassification index (NRI), and integrated discrimination improvement (IDI). The final model 28 included the following independent predictors of PVS: sex, C-reactive protein (CRP) level, 29 interleukin-6 (IL-6) level, the neutrophil-lymphocyte ratio (NLR), monocyte count (MC), 30 albumin (ALB) level, and serum potassium level. The model performed well in both the internal 31 validation (corrected C-statistic = 0.748, corrected Brier score = 0.201) and external validation 32 datasets (corrected C-statistic = 0.793, corrected Brier score = 0.190). The internal calibration 33 was very good (corrected slope = 0.910). The model developed in this study showed high 34 discriminant performance in predicting PVS in nonsevere COVID-19 patients. Because of the 35 availability and accessibility of the model, the nomogram designed in this study could provide a 36 useful prognostic tool for clinicians and medical decision-makers.