Mapping obesity based on neighborhoods was earlier a very cumbersome process involving large amounts of research. But a new Artificial Intelligence developed at the University of Washington promises to make this easier by enabling the AI to spot obesity-prone regions up from space.
AI obesity tracking
The scientists use satellite images of the map and AI to analyze and identify patterns that would indicate the possibility of high obesity in a specific region. Almost 150,000 high-resolution satellite images of six cities — like Seattle, Bellevue, Tacoma, Memphis, Los Angeles, and San Antonio — were used for the study.
“We propose a method for comprehensively assessing the association between adult obesity prevalence and the built environment that involves extracting neighborhood physical features from high-resolution satellite imagery,” Science Alert quotes the AI development team from the paper.
AI then scans the images to identify various types of features, like buildings, trees, parks, roads, water, and so on. It then determines regions that have a higher chance of obesity from an internal set of criteria. For instance, neighborhoods with open green spaces and large parks would indicate that the region is well-off and likely to house people who take good care of themselves. In contrast, tightly packed neighborhoods with little greenery, not much open space, and few parks, would indicate lower physical activity, and therefore higher chance of obesity.
Though the AI system is not 100 percent accurate, it offers an easy to use, highly scalable method to get a rough idea of obesity in specific regions without spending too much money.
“Assessing an area’s obesity rates may help city planners decide whom to encourage to be more physically active or where to make healthy food outlets more prevalent, the paper suggests. Satellite data cannot fully substitute for more traditional measures of public health such as surveys, but as a supplement, it’s cheap and fast. Like a supersize meal,” says an article at Science.
The obesity epidemic in America
The AI system and countless other technical solutions for obesity are being developed for a very simple reason — America is in the middle of an obesity crisis. And unless all available tools are used to analyze and identify solutions to combat the problem, the situation could get much worse. Some people have almost lost hope at making the U.S. a healthy society.
“It’s difficult to be optimistic at this point. The trend of obesity has been steadily increasing in both children and adults despite many public health efforts to improve nutrition and physical activity,” NBC News quotes Dr. Frank Hu, chair of the Department of Nutrition at the Harvard School of Public Health.
A report released by the Centers for Disease Control and Prevention showed that around 20 percent of adolescents and 40 percent of adults in the U.S. were obese. These are the highest obesity rates ever recorded in U.S. history. An area of greater concern is the obesity rates among children — 1 in 10 preschoolers aged 2-5 years, and 1 in 5 kids aged 6-11 years, is estimated to be obese.
If children are not trained to control their appetites, to eat healthy foods, and to do regular exercise, the future obesity figures will be catastrophic. An excessively obese youth will simply suck out the vigor from the U.S. society and send it spiraling downward. As such, the AI obesity identification systems and other technologies are a right step toward gathering data and finding solutions for America’s growing obesity crisis.