Artificial intelligence to evaluate obesity from space



Through a newly developed technique, researchers have used artificial intelligence (AI) technology to estimate obesity on Earth.

They did it by scanning Google Maps images, even without identifying obese people.

The team is led by researchers at the University of Washington in Seattle and has downloaded approximately 150,000 high resolution Google Map images from four American cities for this purpose.

Data on the prevalence of adult obesity were obtained from the Centers for Disease Control (CDC) and Prevention's "500 Cities" project, the study said appeared in the journal JAMA Network Open.

The team then carried these images into a convolutional, in-depth AI network. The process extracted elements from the built environment – the distribution of buildings and green areas.

"The extraction of built environment showed that physical characteristics of a neighborhood (presence of parks, highways, green streets, crossings, different housing types) can be associated with variations in obesity prevalence in different neighborhoods," wrote the researchers.

Obesity has been linked to factors such as genetics, nutrition, physical activity and the environment.

The study stated that convolutional neural networks can be used to automate the extraction of characteristics of the built environment of satellite images for studying health indicators.

"Understanding the association between specific characteristics of the built environment and obesity prevalence can lead to structural changes that can promote physical activity and decline in the prevalence of obesity," the researchers added.

According to the study, access to the obesity rate could help urban planners to develop more green areas to promote more physical activity.






Source link

Leave a Reply