Google has released the COVID-19 Public Forecasts in partnership with the Harvard Global Health Institute. It is a set of models that provide projections of COVID-19 cases, deaths, ICU utilization, ventilator availability, and other metrics over the next 14 days for U.S. counties and states. As reported by VentureBeat, Google said that they will continue to update their model with guidance from its collaborators at Harvard.
Google also added that the COVID-19 Public Forecasts are intended to serve as a resource for first responders in health care, the public sector, and other affected organizations to assist them in preparing for what lies ahead. A way these forecasts can be used is where health care providers could incorporate the forecasted number of cases as a datapoint in resource planning for PPE, staffing, and scheduling. Meanwhile, state and county health departments could use the forecast of infections to assist inform testing strategies and identify areas in danger of outbreaks.
Google said that its researchers developed a completely unique time-series machine learning approach that mixes AI with an ingenious epidemiological foundation to develop the COVID-19 Public Forecasts. By design, the models are trained on public data and leverage an architecture that permits researchers to dive into relationships the models have identified and interpret why they create certain forecasts. They’ve also been evaluated to make sure predictions with reference to people of color — who are hardest hit by COVID-19, with disproportionately high rates of cases and deaths — aren’t wildly skewed or otherwise misleading.
Google claims that their models optimize for high accuracy across all U.S. counties to supply the simplest overall forecast for many communities.
The release of the COVID-19 Public Forecasts follows the launch of Google’s COVID-19 Public Datasets program, which hosts a repository of public data sets concerning the crisis and makes them easier to access and analyze.