Larkin Andrew, Huang Tianhong, Chen Lizhong, Lin Pi-I D, Hart Jaime E, Zhang Wenwen, Coull Brent A, Yi Li, Suel Esra, Hankey Steve, James Peter, Hystad Perry
College of Health, Oregon State University, Corvallis, Oregon 97331, United States.
School of Electrical Engineering and Computer Sciences, Oregon State University, Corvallis, Oregon 97331, United States.
Environ Sci Technol. 2025 Jul 15;59(27):13638-13646. doi: 10.1021/acs.est.5c00966. Epub 2025 Jul 3.
Environmental health studies commonly rely on urban composition measures for built environment exposure assessment. However, quality measures are equally important, as they directly influence health behaviors. We leveraged computer vision and street-view imagery to estimate five components of built environment quality (perceived beauty, relaxation potential, nature quality, safe for walking, and safety from crime) across all U.S. cities, explicitly addressing socio-demographic and temporal biases. We collected 72 516 surveys via Amazon Mechanical Turk, where participants ranked street-view images and provided socio-demographic data. Deep learning models predicted quality metrics at 120 million street locations for 2008, 2012, 2016, and 2020. Cross-validation accuracy ranged from 73% (nature quality) to 59% (safety from crime) compared to 50% expected by random chance. Adjusting sampling weights based on demographics reduced but did not eliminate biases for Hispanic/Latino and Native Hawaiian or Pacific Islander groups (3.5 and 4% lower accuracy, respectively). We also adjusted model predictions for seasonal biases, correcting higher scores from late spring and early summer imagery ( < 0.001). The resulting nationwide estimates of street-level beauty, relaxation, nature quality, and safety for walking (but not safety from crime) can inform epidemiological research, urban planning strategies, and public health interventions.
环境卫生研究通常依赖城市构成指标来评估建成环境暴露情况。然而，质量指标同样重要，因为它们直接影响健康行为。我们利用计算机视觉和街景图像，对美国所有城市建成环境质量的五个组成部分（感知美感、放松潜力、自然质量、步行安全性和免受犯罪侵害的安全性）进行了评估，明确解决了社会人口统计学和时间偏差问题。我们通过亚马逊土耳其机器人收集了72516份调查问卷，参与者对街景图像进行评分并提供社会人口统计学数据。深度学习模型预测了2008年、2012年、2016年和2020年1.2亿个街道位置的质量指标。与随机猜测预期的50%相比，交叉验证准确率在73%（自然质量）至59%（免受犯罪侵害的安全性）之间。根据人口统计学调整抽样权重减少了但并未消除西班牙裔/拉丁裔和夏威夷原住民或太平洋岛民群体的偏差（准确率分别低3.5%和4%）。我们还针对季节性偏差调整了模型预测，校正了春末和夏初图像中的较高分数（<0.001）。由此得出的全国范围内街道层面的美感、放松程度、自然质量和步行安全性（但不包括免受犯罪侵害的安全性）估计值可为流行病学研究、城市规划策略和公共卫生干预措施提供参考。
