A new public forecast using machine learning predicts Wisconsin will confirm more than 15,200 coronavirus cases and 97 deaths between August 1 and 14.
The Covid 19 Public Forecast, developed in a partnership between Google Cloud and the Harvard Global Health Institute, uses historical data and other variables in the SEIR model — tracking the time it takes for an individual to transition between being Susceptible, Exposed, Infected and Recovered — to predict new cases, fatalities and hospitalizations over a 14-day period.
If accurate, the period of the most recent prediction, August 1-14, would represent a slight decrease in deaths over the previous 14-day period; Wisconsin saw 101 people die of coronavirus infection from July 17-31.
It would, however, represent a marked increase in the number of cases confirmed; Wisconsin confirmed 12,433 cases in the two weeks from July 17-31.
The model predicts 13,000 more deaths and more than 953,000 new cases in the United States between August 1 and 14.
The model “is intended to help decision makers in healthcare, the public sector, and other impacted organizations be better prepared for what lies ahead,” according to the model’s User Guide.
A 49-page white paper delves into the statistical modeling and artificial intelligence behind the model.
The model is also searchable by county. A google account is required to view the data.