Researchers using smartphone data to understand how infectious disease could spread

If pandemic flu hit a mid-sized American city, how quickly would it move through the population? Which residents would be most vulnerable to infection? If supplies of vaccines are limited, who should get one?

Researchers at UB and the University of Washington are turning to mobile phone data to inform responses to these crucial public health questions.

The team is mining data from smartphones — which are carried everywhere by their users — to understand how residents of a mid-sized metropolitan area travel around their region and come into contact with one another on a daily basis.

This information will be combined with census data on age and gender to build a computer model that simulates people’s movements and interactions within the area. Then, in the simulation, the scientists will “infect” one or more residents with the flu and track how the disease spreads based on the person-to-person interactions captured by the model.

“Models like these could inform policy-making by showing which communities are more vulnerable, spatially,” says Ling Bian, UB professor of geography and director of the National Center for Geographic Information and Analysis’ UB site. “Models have been around for a long time, but mobile data allows us to approximate people’s behaviors with higher temporal and spatial resolution.”

The project is funded with a $2.7 million grant from the National Institute of General Medical Sciences, part of the National Institutes of Health. The team’s leaders are Bian and Cynthia Chen, University of Washington associate professor of transportation engineering. Enki Yoo, UB associate professor of geography, is also a senior scientist on the team.

Though the focus of the research is Buffalo and its surrounding suburbs, the work represents an advance in modeling that could be used in other cities around the country.

Epidemiological models that simulate the spread of diseases are a key component of emergency preparedness. By identifying which segments of society are vulnerable to infection, such models can help officials make decisions about when and where to issue travel bans, how to administer limited supplies of drugs and vaccines, and where to focus public health campaigns to promote handwashing.