FRIDAY, March 29, 2024 (HealthDay News) — The prevalence of coronary heart disease (CHD) is associated with built environment factors, according to a study published online March 28. It became clear. european heart journal.

Zhuo Chen, Ph.D., of the Harrington Heart and Vascular Institute in Cleveland, and colleagues investigated the association between machine vision-based built environments and the prevalence of CHD, measured using features extracted from Google Street View (GSV). Health outcomes were predicted using convolutional neural networks, linear mixed effects models, and activation maps. A total of 530,000 GSV images were acquired covering 789 census tracts in seven cities in the United States.

The researchers found that 63 percent of the census tract variation in CHD prevalence was predicted by built environment characteristics extracted from the GSV using deep learning. Models that included only composite indicators of census tract-level age, gender, race, income, education, or social determinants of health were improved with the addition of GSV features. Activation maps from features revealed a set of neighborhood features represented by buildings and roads that were associated with CHD prevalence.

“Our study has important implications for the field of clinical practice in health research,” the authors write. “First, we pioneered the use of Street View capabilities in cardiovascular risk assessments, demonstrating a novel approach that introduces a new dimension to our understanding of the impact of the built environment on health.”

Summary/Full text

editorial