Google Cloud has released COVID-19 Public Forecasts in partnership with Harvard Global Health Institute. This is the latest move in the tech industry’s efforts to open up another front in the battle against COVID-19. Do note that the two tech heavyweights — Google and Apple — have already jointly released an Exposure Notification framework and specification for COVID-19 contact tracing.
COVID-19 has thrown economies and societies around the world in disarray as governments work tirelessly to tackle the growing menace. Apart from troubles for the common populace, the front-line workers have been hardest hit as they willingly hold the frontlines. COVID-19 Public Forecasts will be an added resource for first responders in health, public, and allied sectors as the essential workers of our society deal with exposure on a daily basis.
Read More: Linux Foundation Expedites COVID-19 Battle With Notification Projects
What is COVID-19 Public Forecasts?
Google states that the COVID-19 Public Forecasts are trained on public data from institutes such as Johns Hopkins University, Descartes Lab, and the United States Census Bureau. Further, the data will be incessantly updated as per the guidance from the Harvard Global Health Institute. These free forecasts will project COVID-19 cases, deaths, and other metrics for the next 14 days. Currently, this data will be available only for US counties and states.
Speaking about COVID-19 Public Forecasts, Dr. Thomas Tsai, surgeon and health policy researcher in the Department of Surgery at Brigham, stated, “the COVID-19 Public Forecasts model produces forecasts at the critical jurisdiction of public health action—the county. Coupled with the work of the Harvard Global Health Institute’s county-level COVID-19 Suppression Metrics, the COVID-19 Public Forecast Model will allow for targeted testing and public health interventions on a county-by-county basis. By providing accurate, timely predictions of cases, infections, hospitalizations, and deaths to both policymakers and the general public, it will enhance our ability to understand and respond to the rapidly evolving COVID-19 pandemic.”
Read More: Cloud Cost Optimization During and After COVID-19
How Does COVID-19 Public Forecasts Work?
In order to produce these forecasts, Google Cloud is using a novel time-series Machine Learning (ML) approach combining AI with a robust epidemiological foundation. The current model has been trained using public data and comes with an architecture enabling researchers to interpret forecasts as they can study the different relationships learned by the model. The forecasts combined with interpretation by researchers, will vastly help the fight against the pandemic. Further, the company is also releasing datasets used in their white paper and a user guide explaining the full mechanism of Public Forecasts.
Google also stated that COVID-19 Public Forecasts ensure projections related to certain communities in the society hardest hit by COVID-19 — notable people of color — are not skewed or represented disproportionately. In Google’s own words, “we paid close attention to the disproportionate impact the disease has had and how that would impact our adherence to these principles, particularly principle #2: “Avoid creating or reinforcing unfair bias.” COVID-19 Public Forecasts are available for free query in BigQuery and are downloadable as Comma-Separated Value files (CSVs). Further, they are also available via Google’s Data Studio dashboard and its National Response Portal.
Read More: Has COVID-19 expedited Cloud Migration?
What are the Benefits of COVID-19 Public Forecasts?
As already stated, COVID-19 Public Forecasts is aimed as an additional resource for first-responders in health and other organizations. Hence, the forecasts can help the workers better understand and prepare for the pandemic. The forecasts can find multiple uses, and the prominent among them being advanced and contingent future planning of resources.
The forecasted number of cases can help health workers plan and allocate PPE (Personal Protection Equipment), staffing needs, and other scheduling. Further, these forecasts can also help other health officials and administrators identify and assign priority to certain areas projected to be at higher risk. The same 14-day forecasts can also be a helpful tool in planning and implementing testing strategies for COVID-19.
While tackling COVID-19 has been difficult, there can be no doubt that it has indeed brought out the best in our society. We commend Google Cloud for releasing COVID-19 Public Forecasts and its dedication to front-line workers. We hope COVID-19 Public Forecasts helps the society properly tackle the pandemic.