How AI can help fight obesity
Even as much of the globe battles to keep its population well-fed and away from the clutches of malnutrition, some countries in the world face a problem born out of an unhealthy abundance. Obesity has slowly, but surely, gone on to become a major health issue for residents in many parts of the world.
The complex health issue, which is caused by a number of factors, has taken on a life of its own in the last few decades, with the World Health Organisation (WHO) reporting that worldwide obesity has nearly tripled since 1975. As per the information shared by WHO, the situation has gotten so out of hand that by 2016, more than 1.9 billion adults were found to be overweight – of which over 650 million were obese.
Obesity a real threat
Though not acknowledged much, obesity is as serious a threat to human life as malnutrition.
“Overweight and obesity are linked to more deaths worldwide than underweight”, explains WHO. Globally there are more people who are obese than underweight – this occurs in every region except parts of sub-Saharan Africa and Asia.
Using AI to fix obesity
With overweight and obesity becoming such major health risks, a group of US-based researchers has now decided to use AI to help identify and eventually find a fix for the health problem.
Understanding the range of factors that play a part in the rise of obesity – such as the physical and urban environment people live in – the AI tool uses Google Maps images fed to it to estimate the level of obesity in an area.
Interestingly, the software through the course of the process does so without actually searching for obese or overweight people.
How does the tool work?
Though it may sound simple, but the researchers explain, developing a tool sophisticated enough to estimate obesity levels accurately is no easy task.
As per this IANS report, the research only carried out in select neighbourhoods of four US cities – Los Angeles, Memphis, San Antonio and Seattle – required the software to be fed with 1,50,000 high resolution Google Map images of these areas which were then used to extract features of the built environment, such as distribution of buildings and green zones, in the particular areas.
Obesity more than just a genetic problem
Running the collected data through a convolutional deep learning neural network, the researchers realised some interesting things.
For example, obesity, which has long been linked to genetics and diet, was found to have a direct link to the environment and presence of green zones in the neighbourhood.
"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 wrote.
Using AI to build healthier cities
Interestingly, the researchers claim that if the AI tool is rolled out on a larger scale, it can be used to help build healthier neighbourhoods, where problems such as obesity could finally become a thing of the past.
"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," said the researchers.