CDC Selects Delphi Research Group at CMU as Center of Innovation in Outbreak Analytics and Disease Modeling
Center Will Receive $17.5M To Improve Data Use During Public Health Emergencies
By Aaron Aupperlee
Outbreak analytics and disease modeling have become critical tools for guiding decision-making during public health emergencies.
During the COVID-19 pandemic, the Delphi Research Group at Carnegie Mellon University gathered data daily from health care systems, technology companies, medical test results, insurance claims and surveys to help steer policy and public health decisions and change the future of disease and outbreak forecasting.
Now, the Centers for Disease Control and Prevention has selected the Delphi Research Group as a Center of Innovation in Outbreak Analytics and Disease Modeling.
“AI has a remarkable role to play in helping us better anticipate health crises, understand our risks and assess if our interventions are working,” said CMU President Farnam Jahanian. “Carnegie Mellon University’s Delphi Research Group has spent the last decade tracking trends related to the flu and other pathogens and pioneering groundbreaking techniques for epidemic forecasting. The new center will see our researchers and scholars apply the same meticulous lens to new pathogens, expand to new data sources and techniques, and help prepare local and state public health agencies for future public health emergencies."
The CDC’s Center for Forecasting and Outbreak Analytics announced the five-year, $17.5 million funding of the Delphi Center for Innovation in Outbreak Analytics and Disease Modeling on Sept. 19 as part of a national network of 13 centers of researchers, industry leaders and public health departments. The Outbreak Analytics and Disease Modeling Network will work together to create shared resources to address outbreak analytics, and disease modeling and forecasting, and will support a more effective response during public health emergencies.
“Each of the grantees will help us move the nation forward in our efforts to better prepare and respond to infectious disease outbreaks that threaten our families and our communities,” said Dylan George, director of the Center for Forecasting and Outbreak Analytics. “We are committed to working alongside these outstanding partners to achieve our goal of using data and advanced analytics to support decision-makers at every level of government.”
The Delphi Center will work with the other announced centers to build the infrastructure needed to improve readiness and response during public health emergencies. The center will establish connections with state, tribal, local and territorial public health agencies and other local and national partners.
“We are thrilled to join this community and support the important work of the CDC and its Center for Forecasting and Outbreak Analytics. The Delphi Center for Innovation in Outbreak Analytics and Disease Modeling will work to improve on the signals health officials rely on when facing emerging pathogens,” said Peter Jhon, the executive director of the Delphi Research Group and the strategic coordinator of public health research initiatives at CMU. “Forecasting and modeling within a pandemic involves sorting through many data sources. Our work will streamline ingesting as many of these signals as possible and then apply machine learning and AI tools to data to let public health officials know what is happening and what is likely to happen next.”
The Delphi Center will develop, refine and test new outbreak analytics and disease modeling technologies with increased capabilities while developing tools that can assist with public health emergencies in general and specific to a pathogen. To enhance these efforts, the center will work to identify and access new data sources and integrate that new information into forecasting and modeling products and tools. The Delphi Center will collaborate with Optum and other health care industry leaders to increase the level of health information available to researchers and officials while securing patient privacy and anonymity.
“We are honored to support CMU with Optum’s diverse real-world data assets and deep expertise in using data to support the people we serve,” said Michael Eddings, chief executive officer of Optum Serve Technology and Consulting Solutions. “We applaud the CDC’s data-driven efforts to improve public health.”
Finally, the center will expand training, career development and job placement in outbreak analytics and disease modeling with a focus on increased diversity. It will develop courses for post-graduate students and facilitate internships, fellowships and other work experiences for those students with public health agencies and other organizations. This will grow the expertise and capacity for forecasting and modeling within the public health workforce.
"COVID was called a ‘once in a hundred years pandemic,’ but something like it will happen again sooner. We will likely encounter something as challenging again in the next decade," said Roni Rosenfeld, professor and head of the School of Computer Science’s Machine Learning Department (MLD) and co-director of the Delphi Group. "The Delphi Center for Innovation in Outbreak Analytics and Disease Modeling allows us to apply recent lessons learned and develop new strategies to be ready for the next public health emergency, and allows us to build the infrastructure during a time of relative calm. We are thankful to the CDC and its Center for Forecasting and Outbreak Analytics for its support of our research, and we're excited to work within the Outbreak Analytics and Disease Modeling Network.”
The Delphi Research Group was founded at CMU in 2012 to develop the theory and practice of epidemiological tracking and pioneer forecasting techniques to make the technology as universally accepted and useful as weather forecasting. The CMU research group is co-led by co-principal investigators Bryan Wilder, an assistant professor in MLD; and Will Townes, an assistant professor in the Department of Statistics and Data Science at the Dietrich College of Humanities and Social Sciences. Co-principal investigators outside CMU include Ryan Tibshirani, Delphi co-lead and professor of statistics at the University of California, Berkeley; and Daniel McDonald, an associate professor of statistics at the University of British Columbia.
Delphi has been the perennial top performer in CDC-sponsored flu forecasting challenges. In 2019, the CDC tapped the group to establish the National Center of Excellence for Influenza Forecasting. At the start of the COVID-19 pandemic, Delphi shifted its focus to the emerging virus, creating the COVIDcast repository and, in collaboration with Facebook, the COVID-19 Trends and Impact Survey — the largest public health survey ever.