Google Street View predicts heart disease risk based on neighborhood features

Trending 2 weeks ago

NewsGuard 100/100 Score

Researchers person utilized Google Street View to study hundreds of elements of nan built environment, including buildings, greenish spaces, pavements and roads, and really these elements subordinate to each different and power coronary artery illness successful group surviving successful these neighborhoods.

Their findings, published successful nan European Heart Journal today (Thursday), show that these factors tin foretell 63% of nan variety successful nan consequence of coronary bosom illness from 1 area to another.

Coronary bosom disease, wherever a build-up of fatty substances successful nan coronary arteries interrupts nan humor proviso to nan heart, is 1 of nan astir communal forms of cardiovascular disease.

Researchers opportunity that utilizing Google Street View tin thief supply an overview of beingness biology consequence factors successful nan built and earthy environments that could thief not only successful knowing consequence factors successful these environments, but yet thief towards building aliases adapting towns and cities to make them healthier places to live.

The study was led by Prof. Sadeer Al-Kindi and Prof. Sanjay Rajagopalan from University Hospitals Harrington Heart & Vascular Institute and Case Western Reserve University, Ohio, USA, and Dr. Zhuo Chen, a post-doctoral chap successful Prof. Rajagopalan's laboratory.

Prof. Rajagopalan said: "We person ever been willing successful really nan environment, some nan built and earthy environment, influences cardiovascular disease. Here successful America, they opportunity that nan zip codification is simply a amended predictor of bosom illness than moreover individual measures of health. However, to analyse really nan situation influences ample populations successful aggregate cities is nary mean task. Hence, we utilized instrumentality vision-based approaches to measure nan links betwixt nan built situation and coronary bosom illness prevalence successful US cities."

The study included much than half a cardinal Google Street View images of Detroit, Michigan; Kansas City, Missouri; Cleveland, Ohio; Brownsville, Texas; Fremont, California; Bellevue, Washington State; and Denver, Colorado. Researchers besides collected information connected rates of coronary bosom illness according to 'census tracts'. These are areas smaller than a US zip codification that are location to an mean of 4,000 people. The researchers utilized an attack called a convolutional neural network; a type of artificial intelligence that tin admit and analyse patterns successful images to make predictions.

The investigation revealed that features of nan built situation visible connected Google Street View images could foretell 63% of nan variety successful coronary bosom illness betwixt these mini regions of US cities.

Prof. Al-Kindi added: "We besides utilized an attack called attraction mapping, which highlights immoderate of nan important regions successful nan image, to supply a semi-qualitative mentation of immoderate of nan thousands of features that whitethorn beryllium important successful coronary bosom disease. For instance, features for illustration greenish abstraction and walkable roads were associated pinch little risk, while different features, specified arsenic poorly paved roads, were associated pinch higher risk. However, these findings request further investigation.

"We've shown that we tin usage machine imagination approaches to thief place biology factors influencing cardiovascular consequence and this could play a domiciled successful guiding heart-healthy municipality planning. The truth that we tin do this astatine standard is thing that is perfectly unsocial and important for municipality planning."

"With upcoming challenges including ambiance alteration and a shifting demographic, wherever adjacent to 70% of nan world's organization will unrecorded successful municipality environments, location is simply a compelling request to understand municipality environments astatine scale, utilizing machine imagination approaches that tin supply exquisite item astatine an unparalleled level," said Prof. Rajagopalan.

In an accompanying editorial, Dr. Rohan Khera from Yale University School of Medicine, USA said: "The relation of residential location pinch outcomes often supersedes that of known biologic consequence factors. This is often summarised pinch nan look that a person's postal codification is simply a bigger determinant of their wellness than their familial code. However, our expertise to appropriately categorize biology consequence factors has relied connected organization surveys that way wealth, pollution, and organization resources.

"The study by Chen and colleagues presents a caller and much broad information of nan built environment. This activity creatively leverages Google's panoramic street-view imagery that supplements its wide utilized representation application.

"… an AI-enhanced attack to studying nan beingness situation and its relation pinch cardiovascular wellness highlights that crossed our communities, measures of cardiovascular wellness are powerfully encoded successful simply nan ocular quality of our neighborhoods. It is captious to usage this accusation wisely, some successful defining strategical priorities for identifying susceptible communities and successful redoubling efforts to amended cardiovascular wellness successful communities that request it most."

Posted in: Medical Research News | Medical Condition News

Tags: Artificial Intelligence, Blood, Cardiovascular Disease, Climate Change, Coronary Artery Disease, Coronary Heart Disease, Genetic, Heart, Heart Disease, Laboratory, Medicine, Pollution, Research, Supplements, Vascular


While we only usage edited and approved contented for Azthena answers, it whitethorn connected occasions supply incorrect responses. Please corroborate immoderate information provided pinch nan related suppliers or authors. We do not supply aesculapian advice, if you hunt for aesculapian accusation you must ever consult a medical master earlier acting connected immoderate accusation provided.

Your questions, but not your email specifications will beryllium shared with OpenAI and retained for 30 days successful accordance pinch their privateness principles.

Please do not inquire questions that usage delicate aliases confidential information.

Read nan afloat Terms & Conditions.