Google's philanthropic arm Google.org has released a new site that tracks the incidence of flu in the US based on terms used in Google searches.
The system uses aggregated, anonymous results from searches for flu-related terms and plots their locations.
The approach, validated against Centers for Disease Control (CDC) flu records, provides timely data that could be two weeks ahead of government figures.
The site, which is free to use, will pass the early-warning data to the CDC.
Hundreds of billions of Google searches from 2003 onwards were used to develop the model, which was then compared with CDC data on outbreaks.
"Our team found that certain aggregated search queries tend to be very common during flu season each year," Google said in their official blog on the topic.
"We compared these aggregated queries against data provided by the US Centers for Disease Control and Prevention, and we found that there's a very close relationship between the frequency of these search queries and the number of people who are experiencing flu-like symptoms each week."
Traditional survey techniques employed by the CDC take about two weeks to precisely identify outbreaks, and Google hopes that its data, based on a stream of current searches, will serve as an early warning system that the CDC can then act upon.