Google Cloud Releases COVID-19 Public Forecasting Model

August 06, 2020 - Google Cloud and the Harvard Global Health Institute recently launched the COVID-19 Public Forecasts to provide accurate and free projections of coronavirus cases, deaths, and other metrics over two weeks.

The platform will be an additional resource for first responders in healthcare, the public sector, and other impacted organizations to prepare for the future.

“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,” Thomas Tsai, MD, surgeon and health policy researcher at Brigham and Women's Hospital and in the department of health policy and management at Harvard T.H. Chan School of Public Health, said in the announcement.

“By providing accurate, timely predictions of cases, infections, hospitalizations, and deaths to both policy makers and the general public, it will enhance our ability to understand and respond to the rapidly evolving COVID-19 pandemic.”

The COVID-19 Public Forecast leverages public data including data from Johns Hopkins University, Descartes Lab, and the US Census Bureau. The information will continuously be updated with guidance from the Harvard Global Health Institute, Google officials said.

The data from the forecast can be used as resources for those at the front lines of responding to this pandemic who are seeking to better understand and prepare for the progression of COVID-19.

For example, healthcare providers can use the forecasted number of COVID-19 cases as one data point in resource planning for PPE, staffing, and scheduling, while state health departments can use the forecast of infections to help inform their testing strategy and help identify areas at risk of new outbreaks.

“As healthcare providers, the ability to ever more accurately predict the evolution of this pandemic is vital to our ability to prepare for, and manage, the COVID-19 crisis,” said Edmund Jackson, chief data officer at HCA Healthcare.

“Having Google bring their unique compute and AI prowess to better answering this question is enormously helpful. We are excited to be part of this work.”

As part of this initiative, Google Cloud researchers developed a time series machine learning approach that leverages artificial intelligence with an epidemiological foundation.

The model allows researchers to look deeper into different relationships that the model has learned in order to uncover why the COVID-19 Public Forecasts generates the predictions that it does.

“We hope that these measures not only help the public understand how the model works, but can also enable further innovation in infectious disease modeling,” Google officials said.

The COVID-19 Public Forecasts are free to anyone in BigQuery, to download as CSVs, through Google’s Data Studio dashboard, or as part of the National Response Portal.

Separate CDC research uncovered that communities of color in the US have been most affected by the pandemic, with disproportionately high rates of cases and deaths.

Google stated that it is committed to a core set of AI principles, so the company looked to “avoid creating or reinforcing unfair bias.

“Our team has conducted a comprehensive fairness analysis to investigate how that disproportionate impact affects the accuracy of our forecasts and how they should be interpreted,” Google officials said. “We encourage all users who intend to make decisions in part based on the COVID-19 Public Forecasts to closely review the Fairness Analysis.”

Google also reinforced its commitment to help those on the frontlines of the pandemic and urged individuals wanting to learn more about the COVID-19 Public Forecasts to look at the company’s User Guide and White Paper.