AI connects urban design to obesity

Machine learning can process public data in ways that could improve health.
Using satellite imagery, an AI was able to predict where obesity is an issue in thousands of neighbourhoods in cities in the United States. (Pixabay)

As artificial intelligence gets ever more sophisticated, researchers are able to use it to examine issues in detail never possible before.

An AI developed in part part by Elaine Nsoesie, an assistant professor in the Department of Global Health at Boston University, has done that with obesity and the built environment.

Elaine Nsoesie (University of Washington)

The AI combed through tens of thousands of satellite images of urban neighbourhoods, in an attempt to correlate neighbourhood features — parks, stores and sidewalks, for example — with the prevalence of obesity.

Her study was published in the Journal of the American Medical Association Network Open.

Because the AI can process data so quickly and efficiently, it can pick up on hidden patterns that manual analysis might not catch - such as the influence of pet shops, for example.

"What we found were some of these things like pet shops were negatively correlated with obesity prevalence, which could indicate that people in that neighborhood own lots of pets, and might be more likely to go out and walk the pets," Nsoesie explained.

Although there was generally a strong correlation with, say, urban greenery and low obesity, this wasn't always the case. In some dense downtown areas, with almost no parkland but more gyms, there were often lower incidences of obesity as well.

The AI didn't specifically account for the socio-economic factors that often influence health, although "it's possible that the AI is capturing some of those futures if they are present in the environment," she said. She added that the AI might not work as well in different countries, where cultural practices are different.

The next step in the research would be that urban planners could use the AI's findings to help design communities that make it easier for people to live healthier, more active lives, Nsoesie said.

She added that this type of AI-driven research is becoming more common in detecting patterns in issues affecting global health.

"There are several different ways in which we can use those methods to think about health and all the factors that contribute to health."


To encourage thoughtful and respectful conversations, first and last names will appear with each submission to CBC/Radio-Canada's online communities (except in children and youth-oriented communities). Pseudonyms will no longer be permitted.

By submitting a comment, you accept that CBC has the right to reproduce and publish that comment in whole or in part, in any manner CBC chooses. Please note that CBC does not endorse the opinions expressed in comments. Comments on this story are moderated according to our Submission Guidelines. Comments are welcome while open. We reserve the right to close comments at any time.

Become a CBC Member

Join the conversation  Create account

Already have an account?