A Cybernetics-based Dynamic Infection Model for Analyzing SARS-COV-2 Infection Stability and Predicting Uncontrollable Risks

Wenlei Xiao; Qiang Liu; J Huan; Pengpeng Sun; Liuquan Wang; Chenxin Zang; Sanying Zhu; Liansheng Gao

With the study of Wenzhou, it was found that the linkage between Wenzhou and Wuhan could be used to estimate the parameters in Wuhan. At that moment (2020.02.01), the status in Wuhan was extremely confusing, as a lot of infected patients were not confirmed. The hospital beds might be facing a serious shortage. Based on the collected bed data from designated, makeshift, Huoshenshan and Leishenshan hospitals, the model simulated the real situation. What can be confirmed was that emergent quarantining actions should be more strictly taken to admit the infected patients, and much more hospital beds should be prepared to contain the outbreak in Wuhan. Fig. 11 shows the simulated results. We estimated the number of infection based on the confirmed cases from the Japanese and German evacuation. There were 1.4% confirmed cases in the total evacuated population. Fig. 11(a) shows the risk of instability, and Fig. 11(b) gives the result when the forced isolation (namely the subsequent fancang hospitals) was activated since Feb 02, 2020 to curb the spread of the virus. Some days later, the forced isolation in Wuhan was implemented by those temporary shelter hospitals or fangcang. Due to the lack of data, there might exist significant errors in the simulated results. In spite of that, the epidemic stability in Wuhan was successfully estimated, and the model proves that the fancang hospitals played a vitally important role in containing the outbreak at last. The simulation of Wuhan implies a fact that, stability oriented analysis is far more important than accuracy oriented analysis. In order to make a quick decision, administrative responses could activated according to the forecasted risk of instability rather than the forecasted infected numbers. This is because the estimation of stability is usually much faster than the estimation of accuracy.