People in protective masks MOSCOW, 9 Jan. How COVID-19 will spread is trying to predict not only doctors, but also mathematicians. Andrei Leonidov, Doctor of Physical and Mathematical Sciences, said in an interview with Sputnik radio the reason why the latter does not always succeed. The COVID-19 pandemic posed the question of how the infection will spread, when to expect bursts and falls in the incidence. Using the methods of mathematical modeling, at first it was possible to obtain fairly accurate forecasts of the epidemic process, but then the situation became more complicated, said Doctor of Physical and Mathematical Sciences, leading researcher at the P.N. Lebedev Institute of RAS Andrey Leonidov. The rapid evolution of the virus changes many parameters that mathematicians have to take into account. «For example, the time from the virus entering the body to the onset of symptoms when infected with the delta strain is different from previous variants. Even with the original version of the virus, the percentage of seriously ill patients and the percentage of hospitalizations was lower.» — added the expert. For mathematical forecasting to be accurate, it is necessary to update the data on which the model is based at least once a week, he continued. At the same time, scientists should use methods that take into account the influence on the development of the system of each of its changeable elements. «The only way to cope with this is to have a team with an agent-type model (in agent-based modeling, the development of a system is determined by the properties and behavior of all its elements, according to the principle» from the bottom up «, — ed.), Not general, but very specific, in which everything is laid down, and in which parameters critical for prediction would be monitored,» Andrei Leonidov explained in an interview with Sputnik radio. The model should be based on an analysis of dozens of factors, such as the move vaccination, a list of the most dangerous public places in terms of coronavirus infection, and so on, the mathematician concluded.
