Combining machine learning technology with smartphone tracking data to create an application that accurately estimates the spread of the flu.
As suggested by a new study published in the journal Nature Communications , which describes how an application was developed for this purpose .
To create their app, the researchers collected anonymous tracking data from Android phone users in New York City ; Google stores the history of users who have chosen to allow such tracking to be recorded. They used that data to teach a machine learning system to recognize human movement on a city map.
The team then added data from models created to represent flu transmission rates based on patients’ hospital visits and lab reports for the 2016 to 2017 flu season.
They used the app to forecast the spread of flu for the same season . They then compared the results with records from the actual flu season and found that they were as accurate as two of the three conventional systems based on passenger data and better than a third.
Finally, the researchers replicated their efforts to predict the 2016 flu season for all of Australia and found that it could predict the spread of flu in that country with precision .
The researchers note that using phone tracking data is significantly less expensive than using traveler data. They also noted that their system could also be used to track the spread of an outbreak when it crosses international lines, as opposed to systems based on passenger data.