By the Blouin News Technology staff

Tracking viral outbreaks with big data

by in Personal Tech.

A man protects his face while engulfed in a cloud of pesticide during dengue prevention in New Delhi. AFP PHOTO/ Andrew Caballero-Reynolds

A man protects his face while engulfed in a cloud of pesticide during dengue prevention in New Delhi. AFP PHOTO/ Andrew Caballero-Reynolds

The integration of big data analysis into multiple technological arenas is slowly showing itself to be beneficial to not only businesses, city developers, genetic scientists, and government security, but also medical domains. Recent research from the Harvard T.H. Chan School of Public Health in Boston demonstrates how cell phone trackage and widespread data analysis can help researchers pinpoint the outbreak of deadly diseases.

NPR quotes Amy Wesolowski, a Harvard researcher, who noted that the team turned to cell phone data to detect mobility patterns and thus model the spread of Pakistan’s large dengue outbreak in 2013. Two regions were ridden with the disease, and Wesolowski points out that the amount of data the team retrieved would “break a normal computer.”

40 million phone customer accounts were accessed by the team to look at the outbreak. Wesolowski said:

“It’s an emerging infectious disease in Pakistan. Southern Pakistan, near Karachi, has had dengue outbreaks for many years. But in northern Pakistan, around the Punjab region and even farther north, they’ve more recently had dengue outbreaks in the last few years.”

This is not the first use of cell phone data to detect disease outbreak, a tactic experimented with in various parts of Africa. Indeed, at the Blouin Creative Leadership Summit in 2013, thought leaders expounded upon how big data has been used to detect the origin of other nefarious viruses like malaria, using databases consisting of upwards of 15 million data points to figure out where and how the disease spreads.

Separate research, also published in the Proceedings of the National Academy of Sciences in July from Princeton and Harvard groups, used the data from 14,816,512 subscribers in Kenya. The research team compiled data sets comparing mobile phone usage, users’ movement, and the spread of rubella between June 2008 and June 2009. The researchers found that the daily location of all subscribers through tracking their cell communications — having calculated the number of trips subscribers made daily between the country’s then eight provinces — can be linked to the country’s rubella incidence dataset. Over 12 billion mobile phone communications were recorded, according to Princeton.

Lead author C. Jessica Metcalf of Princeton’s Woodrow Wilson School of Public and International Affairs is quoted saying: “One of the unique opportunities of mobile phone data is the ability to understand how travel patterns change over time. And rubella is a well-known seasonal disease that has been hypothesized to be driven by human population dynamics, making it a good system for us to test.”

The long and short of it is that if mobile phone communications can be analyzed broadly to pin and mark where viral outbreaks begin and proliferate, the medical world has a vast new tool in its hands. As with most technologies, big data’s accuracy is in question, but for health officials, it seems only to be a boon.