Alessandro Vespignani: Fighting Diseases in the Age of Big Data
Alessandro Vespignani, The Sternberg Family Distinguished University Professor of Physics, Computer Science and Health Sciences, delivered a talk, “Modeling and Forecast of Contagion Phenomena in the Age of Big Data. photo by Brooks Canaday/Northeastern University

by Joe O’Connell

When it comes to fighting dis­eases in the 21st cen­tury, Big Data is becoming an inte­gral weapon. At the fore­front of this battle is world-​​renowned net­work sci­en­tist Alessandro Vespig­nani, who is using Big Data to pre­dict a pandemic’s progression.

Vespig­nani, the Stern­berg Family Dis­tin­guished Uni­ver­sity Pro­fessor of Physics, Com­puter Sci­ence, and Health Sci­ences at North­eastern, recently dis­cussed the ways his MoBS Lab is using all types of data—from flight pat­terns to mobile phone information—to deter­mine the spread of a disease.

His talk on Monday after­noon, which was held in the Raytheon Amphithe­ater, was part of the “Minds over Mat­ters: NUterm Fac­ulty Speaker Series.” The series fea­tures weekly pre­sen­ta­tions from top fac­ulty scholars who dis­cuss their research and examine inno­va­tion, new dis­cov­eries, and timely topics of global importance.

Here are five things we learned from Vespignani’s lecture:

Get on a micro­scopic level

The best way to develop a model that projects future trends, Vespig­nani explained, is to look at a system on the micro­scopic scale. By fol­lowing a single individual—where she goes during the day, how she gets there, and with whom she inter­acts —Vespig­nani and fellow net­work sci­en­tists can create a large-​​scale sim­u­la­tion of the evo­lu­tion of a dis­ease.

Trans­porta­tion has greatly changed how dis­ease is spread

When the Black Death was spreading through Europe in the 14th cen­tury, its only means of mobility was via infected people trav­eling to nearby vil­lages for trade. That, Vespig­nani said, cre­ated a “wave” showing how the dis­ease spread at dif­ferent veloc­i­ties based on where people traveled.

Nowa­days, due to planes, trains, and other modes of trans­porta­tion, epi­demics like the SARS out­break in the early 2000s travel much quicker and in point-​​to-​​point fluxes.

Behav­ioral change is a major hurdle

Human behavior, Vespig­nani said, is a sig­nif­i­cant bar­rier to devel­oping accu­rate pre­dic­tion models. Once a dis­ease starts to spread, humans change their day-​​to-​​day behavior to avoid infec­tion, which impacts the accu­racy of predictions.

In a place like the U.S., the behavior would change so dras­ti­cally that our model would prob­ably struggle to get some real­istic feed­back in the system,” Vespig­nani explained. “You have this loop you need to close in which the pre­dic­tive system is part of the effects you need to predict.”

More than just pre­dicting diseases

To close his talk, Vespig­nani explained that his lab’s fore­casting model can be used for other sys­tems as well. In today’s society, he said, ideas and polit­ical pref­er­ences can spread through social media just like dis­eases spread through nature.

Every­thing has a spreading process,” Vespig­nani noted, “and when we look at the con­ta­gion process, rather than fol­lowing an indi­vidual we follow an event.”

Culling from Big Data

During a Q&A with the audi­ence, Vespig­nani was asked to describe his strategy for culling data from the moun­tains of avail­able infor­ma­tion. He noted that it depends on the ques­tions the sci­en­tist is asking. “There is no recipe,” he said. “You really have to figure it out case by case.”

Originally published in news@Northeastern on May 20, 2015.