R-CCS Cafe 第205回 第1部

In 2020, the COVID-19 pandemic has deeply impacted the world. Researchers from various fields have been working on understanding and fighting back against the disease. For the Public Health Service, tracking the effective reproductive number R_t, fitting the mathematical models to the observations, and making good predictions and simulations are important. However, a unique feature of COVID-19 brings a large uncertainty in their estimates and predictions: namely, the existence of a large population of transmissive infections without symptoms. Detecting the asymptomatic individuals is not easy, and the information provided by the Public Health Service on asymptomatic cases is limited. However, asymptomatic individuals can easily become super-spreaders, and estimating this population is of crucial importance. In this talk, I introduce an extension to a well-known Susceptible-Infectious-Recovered (SIR) model and show how we apply the ensemble Kalman filter for estimating the undetected infected individuals and other relevant parameters.