Abstract: Objective To establish the risk nomogram model for norovirus (NoV) infection in children through clinical symptoms and routine blood biochemical indicators, and verify the predictive efficacy, so as to provide a reference for early and rapid screening of NoV infection in clinical practice. Method A total of 412 children (3 months to 16 years old) diagnosed with acute viral gastroenteritis in the Third People's Hospital of Zhangjiagang City from January 2023 to October 2024 were retrospectively included as a training set, and were randomly divided into a modeling group (330 cases) and an internal validation group (82 cases) at a ratio of 4:1. In addition, 202 children with acute viral gastroenteritis from the Children's Hospital of Soochow University between October 2023 to June 2024 were selected as the external validation group. Real-time quantitative PCR (rq-PCR) was used to confirm NoV infection. The clinical symptoms and blood biochemical indicators were compared between NoV-positive and NoV-negative children in the modeling group. Multivariate Logistic regression was used to screen predictors and construct a nomogram model. The receiver operating characteristic curve was used to evaluate the predictive efficacy of the model in the internal validation group (n=82) and external validation group (n=202). Result Single-factor comparison showed that white blood cell count (WBC), absolute counts of neutrophils and lymphocytes (NE and LY), and platelet count (PLT) in positive children were significantly higher than those in negative children,vomiting, diarrhea, and upper respiratory tract infection symptoms were were more common, and clinical symptoms and treatment methods were more complicated as well, too(P<0.05). Multivariate logistic regression analysis showed that vomiting (OR=1.859, 95%CI=1.325-2.253, P=0.001), upper respiratory tract infection (OR=1.234, 95%CI=1.052-1.564, P=0.007), WBC (OR=2.325, 95%CI=1.857-2.659, P<0.001), LY ≥3.42×109/L (OR=1.425, 95%CI=1.124-1.865, P=0.003) and PLT (OR=3.023, 95%CI=2.425-3.756, P<0.001) were the predictive factors ot NoV infection. The nomogram model constructed based on the above factors had a total score of 220. In the internal validation group, the area under the curve (AUC) for predicting NoV infection was 0.842 (95%CI=0.768-0.899, P<0.001) with a sensitivity of 78.9% and specificity of 75.0%. In the external validation group, the AUC was 0.896 (95%CI=0.845-0.943, P<0.001) with a sensitivity of 83.3% and specificity of 81.9%. Conclusion The nomogram model established based on clinical symptoms and routine blood biochemistry of children with acute viral gastroenteritis shows good potential for diagnosing NoV infection, and can assist in rapid and accurate clinical screening of NoV.

Key words: Norovirus, Acute viral gastroenteritis, Risk factors, Nomogram, Clinical symptoms, Blood biochemical indicators