Mobile phones can help predict epidemics

[Shiloh]

Source: http://www.telegraph.co.uk/earth/mai...ciphone104.xml

Mobile phones can help predict epidemics

By Roger Highfield, Science Editor
Last Updated: 6:01pm BST 04/06/2008

The plague of mobile phones that has spread across the planet could become a vital tool to help predict the next influenza pandemic by revealing the mass movements of people.

# Vaccine to combat all types of flu 'by 2011'
# Larry Brilliant, of Google.org: Internet 'is pandemic early warning system'

Governments around the world are preparing for the next pandemic, when an avian strain of flu virus develops the means to spread among people and kill millions worldwide.

Now critical information on how to anticipate its spread will come from the mobile phone, thanks to a study by Prof Albert-László Barabási, Marta González, and César Hidalgo at Northeastern University, Boston.

Mobile phones - great for fixing up meeting places and letting people know you're running late - have a new use: tracking patterns of human movement, providing crucial information for modelling the spread of disease, among other things.

A new study, published by the team in the journal Nature, has monitored the movements of 100,000 people by their mobile-phone signals, and perhaps unsurprisingly shows that most people are creatures of habit, tending to make regular migrations to the same few locations, but with occasional long hops.

Mapping large-scale human movements is important for urban planning, traffic forecasting and disease monitoring, but previous attempts to model these patterns, such as by tracking bank notes, have not provided an accurate picture of individuals' movement.

"The quantitative understanding of human dynamics is one of the most under-explored areas of contemporary science," Prof Barabási says. "Mobile phone activity is a rich set of information because its high rate of penetration in the society, nearly 100 per cent in some countries."

The data can also suggests how to monitor the way people behave in emergencies in real time, and locate where an emergency is taking place. For example, a pile up on a motorway causes lots of rapidly moving phones grind to a halt, a few call the emergency services while others call the office/spouse/lovers.

"The study also showed that there is a great deal of variation in the average distance travelled by different individuals. While 73 per cent of the people spend most of their time within a 10 mile radius, there is a 3 per cent of the population that moves regularly over more than 100 miles," says Hidalgo.

The patterns revealed in the new study differ slightly from the classic "Levy flight" patterns displayed by many foraging animals, which are much more random, probably because most humans are not completely free to roam but instead have to turn up to work every day.

While it is likely that the wanderings of one albatross could be similar to that of another, "it is not the case of humans," says Prof Barabási.

Higher recurrence in the visited places is also observed for humans, which is not necessarily the case for animals, that may tend to move to further locations once food is expired."

He adds: "we are not at liberty to disclose the data's origin.

[tropical]

Because this kind of tracking is already part of the phones, this is a good initiative.

Looking at 100 years ago, when the track was a horse footprint, or a gossip, tracking mobcells positions is an enormous intruding step in peoples private life habits.