Cpx24.com CPM Program

Wednesday, November 12, 2008

Google Flu - Google search engine flags flu activity


Search engine giant Google launched a new tool on Tuesday that will help U.S. federal health experts track the annual flu epidemic.

Google Flu Trends uses search terms that people put into the Web-based search engine to figure out where influenza is heating up, and notify the U.S. Centers for Disease Control and Prevention in real time.

"We've discovered that certain search terms are good indicators of flu activity," Google said in a statement.

"What this does is it takes Google search terms of influenza-like illness and influenza and it emulates a signal that tells us how much influenza activity there was," Dr. Lyn Finelli, chief of influenza surveillance at the CDC, said in a telephone interview.

Studies indicate that between 35 and 40 percent of all visits to the Internet are begun by people looking for health information. When people are sick, they tend to look up their symptoms.

Google is keeping the search terms it uses private, but influenza-like illnesses include symptoms such as fever, muscle aches and cough. Sneezing usually occurs with other viruses such as rhinoviruses.

Currently, the CDC relies on centers that report on people coming to their doctors with flu-like symptoms, and lab tests that confirm whether a patient has influenza.

But many people with flu never visit a doctor and most doctors treat based on symptoms, rarely giving a flu test.

Either way, the CDC's surveillance data is about two weeks behind.

The Google tool will track flu activity in near real time, the company said.

EARLY WARNING

"One thing we found last year when we validated this model is it tended to predict surveillance data," Finelli said.

"The data are really, really timely. They were able to tell us on a day-to-day basis the relative direction of flu activity for a given area. They were about a week ahead of us. They could be used ... as early warning signal for flu activity."

Then the CDC can get the word out to hospitals, clinics and doctors offices so they can stock up on flu tests, antiviral drugs and antibiotics for people who get what are known as co-infections -- bacterial infections that worsen a bout of flu.

Two weeks warning also allows people to get vaccinated before flu reaches their community.

Google Uses Searches to Track Flu’s Spread

Now it seems Big Brother will know you're sick even before you do.

Google has launched a function called Google Flu Trends, which will track which areas of the country are punching in flu-related search terms in a bid to track the spread of the disease.

"Our team found that certain aggregated search queries tend to be very common during flu season each year," the Web site posted on its official blog.

Terms such as "flu symptoms" or "muscle aches" are some common examples.

"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," the company said.

The flu season has not yet gotten under way. Most states are reporting low flu activity, except for Nevada and Texas, where the activity is deemed minimal. Flu cases typically spike in mid-December and again in February.

The search engine touts its new data page as "an early warning system for outbreaks," making it "an exciting development" for epidemiologists.

"Our up-to-date influenza estimates may enable public health officials and health professionals to better respond to seasonal epidemics and - though we hope never to find out - pandemics," the company said.

As part of a test to measure how accurate the data they had collected is, Google researchers passed their results to the CDC throughout last winter's flu season.

They found that the search terms had a very "strong correlation" to what health officials were seeing, except that their data was coming in sooner. That provided the possibility to detect outbreaks as they were happening.

So far, however, Google acknowledges the patterning is preliminary and experimental.

"Our system is still very experimental, so anything is possible, but we're hoping to see similar correlations in the coming year," Google said.

Google also promised the search patterns would never be connected to personal data.

No comments:

Post a Comment