Quantcast
NYU researchers use algorithms to predict flu outbreaks - Metro US

NYU researchers use algorithms to predict flu outbreaks

nyu researchers predict flu outbreaks computer model
Social interactions come in bursts, NYU researchers argue, and can't be modeled like constant forces.
Getty Images

Affecting more than seven million people across America in this year alone, knowing when and how outbreaks of the disease can help New York hospitals prepare for influxes of patients with the potentially life-threatening disease. Researchers at NYU have developed a tool to do just that–to predict flu outbreaks by simulating how Saturday brunches to Sunday masses affect the spread of disease.

“We all know that human contact plays an important role in the spread of disease, but up until now we haven’t had a good way to predict its effect — both before and during an outbreak,” said Maurizio Porfiri, a professor of Mechanical and Aerospace Engineering, and Biomedical Engineering.

Relying on “precise mathematical equations and algorithms,” the techniques in Porfiri’s paper, published in the SIAM Journal on Applied Dynamical Systems, argue that previous flu simulations can be improved by measuring not just where people go for work, but how they spend their free time.

“Social scientists have already proven that human interaction occurs in bursts of activity,” Porfiri explained. “Our computer model accurately predicts when those bursts will happen over time, helping to paint a better picture of how quickly disease is likely to spread or die down.”

Researchers at Columbia, just last year, developed tools to predict flu outbreaks in the same way as weather forecasts, by using information on where and when people commute for work. In any case, Porfiri argues that the CDC can use their research to better prepare for flu season, and to provide areas with the vaccines that they need more accurately and ahead of time.

“We know that flu season is going to arrive each year,” said Lorenzo Zino, another of the scientists involved with the project. “By providing more accurate information that includes how people will behave, our tool can help to reduce the number of people affected.”

More from our Sister Sites