Home » News and Events » AI can speculate an area’s obesity level from its infrastructure

AI can speculate an area’s obesity level from its infrastructure

An AI algorithm is developed to estimate a city’s obesity rate by analyzing satellite images of its buildings.

 

A research group at the University of Washington analyzed images from Google Maps of four cities in U.S (Los Angeles, California; Memphis, Tennessee; San Antonio, Texas; and Seattle, Washington) to draw out the prediction for the rate of obesity–an alarming health issue in this country.

The team found the relationship between an area’s built environment and its neighborhood’s health issues by using neural networks. The images of the urban areas provide health indicators to the neighborhood ass the presence of parks, highways, green streets and housing types can be associated with the people’s physical activity according to a research published by JAMA Network Open.

Moreover, the algorithm can use neural networks to automate the extraction of features of infrastructure from satellite images. This study can possibly result in adjustment in infrastructure  to encourage physical movement and increase people’s health.