0
0
In the digital age, big data and artificial intelligence (AI) are transforming public health research and policy. A new study from New York University (NYU) suggests that even Google Street View images can offer valuable insights into public health, though they need to be used alongside other data to maximize their potential. This research, published in the Proceedings of the National Academy of Sciences, evaluated two million Google Street View images from New York City to understand how they could help predict health outcomes such as obesity and diabetes.
The NYU researchers, led by Rumi Chunara, Associate Professor at NYU’s School of Global Public Health, explored the connections between a city’s built environment—specifically sidewalks and crosswalks—and public health. However, they discovered that relying on street view data alone can lead to inaccuracies, underscoring the importance of blending this information with expert knowledge.
Street-Level Insights Into Health
Linking environmental factors like neighborhood infrastructure to health outcomes such as mental health, infectious diseases, and obesity is an area of growing interest. For instance, many studies have previously shown that sidewalks are correlated with lower obesity rates. However, the NYU study found that not all sidewalks are created equal, and simply having them may not always translate to better health outcomes.
“There’s a lot of excitement around leveraging new data sources to gain a holistic view of health, including bringing in machine learning and data science methods to extract new insights,” said Chunara. “But it’s important to recognize the limitations of such data, particularly when it comes to the complex relationships between environment, behavior, and health outcomes.”
The team used AI to assess the availability of sidewalks and crosswalks in the street view images, comparing the data to health statistics from the Centers for Disease Control and Prevention (CDC). While neighborhoods with more crosswalks showed lower rates of obesity and diabetes, sidewalks did not consistently predict better health. In many cases, sidewalks existed in areas where people were less likely to walk, such as highways or bridges.
The Complex Relationship Between Built Environment and Health
The study emphasized that while infrastructure like crosswalks might contribute to health improvements, other factors—like physical activity—play a much larger role. For example, increasing physical activity had a significantly larger impact on reducing obesity and diabetes than simply adding more crosswalks. This suggests that interventions aimed at promoting active lifestyles may be more effective in improving public health outcomes than focusing solely on urban infrastructure.
“Physical activity delivers the benefits of crosswalks, so it’s important to take such mechanisms into account,” said Miao Zhang, the study’s lead author and a Ph.D. student at NYU Tandon School of Engineering. “Especially when they act on different levels, like the built environment versus individual behavior.”
The Future of Public Health Decision-Making
This study highlights a key takeaway for public health experts: while digital data, such as street view images, can provide useful insights, they should not be used in isolation. Incorporating domain knowledge—whether from urban planning, epidemiology, or computer science—into the analysis is essential for making informed, effective public health decisions. For example, understanding how AI image processing works can help correct for bias or mislabeling in street view images, improving the accuracy of public health predictions.
As cities and governments look to use data-driven approaches to tackle public health challenges, this research advocates for a more nuanced strategy. Instead of focusing on superficial associations between infrastructure and health, public health programs should consider how various factors—like physical activity and community engagement—interact with the built environment.
“While growing amounts of digital data can be useful in informing decision-making, our results show that simply using associations from new data sources may not lead to the most useful interventions or best allocation of resources,” Chunara concluded. “A more nuanced approach using big data in conjunction with expertise is needed to make the best use of this new data.”
This groundbreaking research highlights the potential—and the challenges—of using digital data in public health. The future of health policy may lie in such innovative, data-driven approaches, but experts warn that without a deep understanding of the data’s limitations, interventions could miss the mark.
References:
Zhang, M., Chunara, R., et al. “Utilizing Big Data Without Domain Knowledge Impacts Public Health Decision-Making,” Proceedings of the National Academy of Sciences (2024). DOI: 10.1073/pnas.2402387121.
