AI WILL PREDICT A CITY’S FATNESS LEVELS FROM ITS BUILDINGS

AI will estimate fatness rates by analysing satellite imaging of urban areas

An artificial intelligence algorithmic program has been trained to identify cities at risk of fatness just by analysing its buildings and infrastructure.

Researchers at the University of Washington used satellite and Street read pictures from Google Maps to check four urban areas within the United States: la, California; Memphis, Tennessee; urban center, Texas; and city, Washington.

Physical characteristics of a neighborhood, such as parks and high-rise buildings, can have a significant impact on its inhabitants' health

Using neural networks, the researchers were able to discover a relationship between a location’s infrastructure and its fatness prevalence.

Obesity could be a major health issue within the U.S.A., with one in 3 adults plagued by the condition. various studies are distributed to higher perceive the development, with previous analysis linking fatness to factors as well as biological science, diet, physical activity and also the setting.

The most recent study was distinctive within the tools it accustomed examine the link between engineered environments and fatness levels.

“This study illustrates that convolutional neural networks may be accustomed modify the extraction of options of the engineered setting from satellite pictures for finding out healthy indicators,” wrote the authors of the study, that was revealed in JAMA Network Open.

“Results within the study support the association between options of the engineered setting and fatness prevalence.”

Further analysis may specialise in assessing disparities supported neighbourhood racial composition and socioeconomic standing, that has been connected to fatness and alternative health problems.

“Care should be taken in not over-interpreting any results,” aforesaid biostatistician Benjamin Goldstein from university, during a comment on the study. “Even so, within the same manner a biomarker might function a helpful indicator of illness risk, these neighbourhood factors will function a valuable indicator of health outcomes.

“Going forward, it’s probably that machine learning ways are going to be integral to discovering options related to illness – probably options ne’er antecedently suspected.”

Eventually, the analysis may facilitate public health bodies and improve community designing efforts for the aim of fighting fatness.

The study’s authors say the results might be accustomed build structural changes to engineered environments at risk of fatness so as to encourage physical activity and introduce alternative measures to push healthy living.

The study concludes: “Neighbourhood-level interventions to encourage physical activity and increase access to healthy food shops might be combined with individual-level interventions to assist in curb the fatness epidemic.”

Advertisements

Leave a Reply

Your email address will not be published. Required fields are marked *