Do you think the presence of gyms, parks, green streets, and swimming pools in your neighbourhood could be linked to your physical fitness? Or the fact that you are surrounded by fast food outlets, highways or convenience stores can have a negative effect on your health?
The researchers, from the University of Washington, thought along the same lines and used artificial intelligence (AI) and satellite images of US cities to track adult obesity levels from space.
Obesity has commonly been linked to factors such as genetics, diet, physical activity, and the environment.
“We propose a method for comprehensively assessing the association between adult obesity prevalence and the built environment that involves extracting neighbourhood physical features from high-resolution satellite imagery,” the team explained in a paper.
The researchers fed some 150,000 high-resolution satellite images from Google Static Maps API (application programming interface) into a previously trained convolutional neural network (CNN). CNN is a deep learning approach to independently analyse and identify patterns within the dataset.
The data covered 1,695 census tracts to extract features of the built environment in: Bellevue, Seattle, Tacoma, Los Angeles, Memphis, and San Antonio.
“The extraction of built environment showed that physical characteristics of a neighbourhood (presence of parks, highways, green streets, crosswalks, diverse housing types) can be associated with variations in obesity prevalence across different neighbourhoods,” the researchers added.
The neural network used by the team in this case was pre-trained on approximately 1.2 million images. The idea was to study the efficacy of CNNs at associating obesity prevalence with significant physical environment features.
As the findings reveal, the built environment can have a direct influence on people’s health through the following resources:
• Activity and recreational spaces
• Measures of community design
The study unfolds possibilities for improving the methods for a more consistent application, according to the researchers.
Understanding the association between specific features of the built environment and obesity prevalence can lead to structural changes that could encourage physical activity and decreases in obesity prevalence, the study concluded.