Abstract：Objective To analyze the risk of imported COVID-19 cases in Xiamen, and to provide data support for the public health department to prevent the spread of imported cases; to assess the future shortage of medical resources in Xiamen, and to provide a reference basis for administrative decisions.Methods A statistical model was constructed to predict imported cases in Xiamen. Due to the small number of initial samples in Xiamen, the model was first initialized based on the overseas import situation of Guangzhou, and then fitted based on the daily epidemic situation in Xiamen. Considering the constant change of entry policies and the rapid development of epidemic in various countries (regions), a dictionary was constructed and maintained to dynamically calculate the risk weights of travelers from different countries. Finally, exponential smoothing method was used to update the parameters of the model.Results From March 23, 2020 to December 31, 2020, Xiamen reported a total of 277 imported cases in which asymptomatic infections were included, and 257 cases were predicted by our model. The experimental results showed that the F1-score of our proposed model reached 77.3%.Conclusion The statistical model has good predictive-ability.
张怡盾，童逸琦，黄仕杰，黄思颖，庄福振. 厦门市新型冠状病毒肺炎境外输入确诊人数的预测模型构建和评价[J]. 中国医院统计, 2021, 28(6): 513-517.
Zhang Yidun, Tong Yiqi, Huang Shijie, Huang Siying, Zhuang Fuzhen. Construction and evaluation of a statistical model for COVID-19 imported cases in Xiamen. journal1, 2021, 28(6): 513-517.
［4］YANG Z, ZENG Z, WANG K, et al. Modified SEIR and AI prediction of the epidemics trend of COVID-19 in China under public health interventions［J］. J Thorac Dis, 2020, 12(3):165-174.
［5］WANGPING J, KE H, YANG S, et al. Extended SIR prediction of the epidemics trend of COVID-19 in Italy and compared with Hunan, China［J］. Front Med (Lausanne), 2020, 7:169.