Method and Application of Estimating Epidemiological Parameters Based on Data-Driven Approach

Abstract

Determining initial variables and key parameters, such as case fatality ratio (CFR), dynamic case fatality ratio (DCFR), reproduction number ( $R_{0}$ ), and so on, helps shed more light on the transmission and control of emerging and re-emerging infectious diseases. Here, we established a SAIUHR model, which describes the dynamic changes of susceptible, asymptomatic infectious, under-reported symptomatic infectious, hospitalized and recovered individuals. And we proposed a novel approach based on our model to calculate the report rate, starting time, basic reproduction number, the initial conditions for the compartments, CFR and DCFR. Finally, we apply our method to epidemiological datasets from China, Italy, Germany, and France. The results show that the goodness of fit for the cumulative confirmed cases is greater than 97.45% in each of the countries, DCFR is more effective than CFR in predicting the future tend of infectious disease, and improving the report rate, raising the control strength and shortening the wait time are the effective strategies against infectious diseases. This study highlights the implications of taking proper restrictions and strong policies to deal with emerging and re-emerging infectious diseases from their spread in the early stage.

Keywords

• Basic reproduction number
• data-driven
• epidemic model
• parameter estimation