To better prepare Minnesota for the impact of COVID-19 on the state, the University of Minnesota School of Public Health and Minnesota Department of Health (MDH) joined forces in mid-March, 2020, to build a mathematical model predicting the progression of the disease and its impact on Minnesota.
To run the model’s equations, create its graphic elements, and produce various scenarios, the team wrote specific code using the R language and software.
More on the Minnesota Model
To the classic equation for modeling infectious diseases called SEIR (Susceptible, Exposed, Infectious, Recovered), the first two versions of the Minnesota model added three extra stages representing people who are: Hospitalized, in ICU, or Dead. Its third version, released on May 8, 2020 [add link], incorporates emerging U.S. and Minnesota data about the pandemic and adds a new element to the equation: AI for Asymptomatic Infection.
Using the R code, the model can run different scenarios showing how people may move in and out of some of the stages over time and it can produce graphs, charts, and tables to make the information visible and accessible. This flexibility allows public health officials to try out different social distancing measures to predict how those actions affect the timing and number of COVID-19 cases and deaths in the state, and how soon our health care systems would be overloaded.
Originally, the team used estimates from China and Europe to craft some parameters for the Minnesota model because U.S. data was still limited. But the model is becoming ever more state specific as U.S. data grows and Minnesota-centric information becomes more robust. The team plans to produce new model versions every _______ weeks.