Coronavirus disease 2019 epidemic trend prediction based on improved infectious disease dynamics susceptible-exposed-infected-recovered (SEIR) model / 第二军医大学学报

ABSTRACT

Objective To predict the coronavirus disease 2019 (COVID-19) epidemic situation based on the infectious disease dynamics susceptible-exposed-infected-recovered (SEIR) model, so as to provide guidance for effective control of the epidemic. Methods Python crawler automatic update function was used to collect the epidemic data released by the National Health Commission of China. An improved infectious disease dynamics SEIR model, which can automatically correct the COVID-19 basic reproductive number (R0), was constructed to predict the development trend of COVID-19 epidemic in Hubei Province of China and South Korea. Results The peak of the COVID-19 epidemic in Hubei Province of China predicted by the model would appear on Feb. 21, 2020. The number of confirmed COVID-19 cases would be about 50 000 on Feb. 19 and would fall to below 30 000 on Mar. 4, and the epidemic would end on May 10. According to the actual data released by the National Health Commission of China, the peak number of confirmed COVID-19 patients was 53 371. The model predicted that an epidemic peak in South Korea would be on Mar. 7, and would end at the end of April. Conclusion This improved infectious disease dynamics SEIR model established in the early stage of COVID-19 epidemic has achieved relatively accurate prediction. The timely and effective intervention by relevant government departments has significantly affected the development of the epidemic. The epidemic situation in other countries in East Asia, such as South Korea, is still on the rise in March, suggesting that China needs to be on guard against the risk of imported epidemic.