Two weeks after beginning to question users about signs of the disease with an initiative to monitor the spread of COVID-19, Facebook said on Monday that early findings showed promise — and that the project would be spreading out globally beginning Wednesday. Carnegie Mellon University, which is undertaking the study, said 150,000 people a day now experience symptoms after clicking on a link in the News Stream. The university is not posting any symptom information back to Facebook.
In an op-ed in the Washington Post, Facebook CEO Mark Zuckerberg said that Carnegie Mellon’s initial results align with public data regarding reported COVID-19 outbreaks, a significant indication that symptom surveys will likely be able to predict the spread of the disease. Using aggregated university info, Facebook will generate interactive maps based on symptom surveys and update them regularly.
“I have always felt that seeing people come together as a group would help us overcome our toughest challenges — not only by sharing our stories and encouraging each other in emergencies but also by working together to solve issues on a global scale,” Zuckerberg wrote. “The world has faced pandemics before but we have a different superpower this time: the opportunity to gather and exchange data for good. I am hopeful that, if we use it wisely, data will help the world adapt to this health crisis and get us on the road to recovery.
Facebook has announced the collaboration with Maryland University to start gathering data on the symptoms globally. If Facebook users experience symptoms at levels close to their U.S. presence in their own countries, the data may prove to be an accurate map of possible hotspots across the globe, said Ryan Tibshirani, an associate professor at Carnegie Mellon who helps lead the school’s relationship with Facebook.
Also Read: COVID-19 Tracker App “Rt.Live” Releases by Instagram
As the pandemic has spread across the globe, Zuckerberg has continued to illustrate how the vast social presence of Facebook can be used to help political leaders and public health authorities make more educated choices about their pandemic response. Last month, Facebook started integrating a “COVID-19 Data Center” into the News Feed, showcasing users approved public health authority data. To date, the center has served more than 350 million people, Zuckerberg said.
In a separate statement on Monday, Facebook revealed new collaborations with academics to use machine learning methods to make more predictions. In New York, the organization partners with New York University and Cornell University to build forecasts for disease transmission at the county level, as well as projections about how an uptick in cases would impact demand for medications, ventilators, and personal protective equipment in hospitals in the region. Another method would use Facebook’s machine learning algorithms to evaluate de-identified X-rays and computerized tomography (CT) scans with a view to forecasting medical conditions and demand for equipment.
Zuckerberg said that while Facebook does not try to analyze the data on the symptoms it shares with researchers, its presence has allowed it to make a substantial contribution to the response to public health.