# Using Deep Learning to Examine the Association between the Built Environment and Neighborhood Adult Obesity Prevalence

arXiv: https://arxiv.org/abs/1711.00885 · HTML: https://mlsanity.com/paper/1711.00885/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Nov 2017.

## Built on by (most cited first, 8 of 8)

- [SustainBench: Benchmarks for Monitoring the Sustainable Development Goals with Machine Learning](https://mlsanity.com/paper/2111.04724/) (2021)
- [Geospatial Foundation Models to Enable Progress on Sustainable Development Goals](https://mlsanity.com/paper/2505.24528/) (2025)
- [Jane Jacobs in the Sky: Predicting Urban Vitality with Open Satellite Data](https://mlsanity.com/paper/2102.00848/) (2021)
- [CureGraph: Contrastive Multi-Modal Graph Representation Learning for Urban Living Circle Health Profiling and Prediction](https://mlsanity.com/paper/2501.07157/) (2025)
- [Supervised Machine Learning for Breast Cancer Risk Factors Analysis and Survival Prediction](https://mlsanity.com/paper/2304.07299/) (2023)
- [Can Machine Learning Uncover Insights into Vehicle Travel Demand from Our Built Environment?](https://mlsanity.com/paper/2311.06321/) (2023)
- [Observing Health Outcomes Using Remote Sensing Imagery and Geo-Context Guided Visual Transformer](https://mlsanity.com/paper/2602.00110/) (2026)
- [Geospatial Foundation Models Capture Health-Relevant Dimensions of Place Beyond Conventional Social Risk Indices](https://mlsanity.com/paper/2609.11689/) (2026)

## Abstract (arXiv)

More than one-third of the adult population in the United States is obese. Obesity has been linked to factors such as, genetics, diet, physical activity and the environment. However, evidence indicating associations between the built environment and obesity has varied across studies and geographical contexts. Here, we used deep learning and approximately 150,000 high resolution satellite images to extract features of the built environment. We then developed linear regression models to consistently quantify the association between the extracted features and obesity prevalence at the census tract level for six cities in the United States. The extracted features of the built environment explained 72% to 90% of the variation in obesity prevalence across cities. Outof-sample predictions were considerably high with correlations greater than 80% between predicted and true obesity prevalence across all census tracts. This study supports a strong association between the built environment and obesity prevalence. Additionally, it also illustrates that features of the built environment extracted from satellite images can be useful for studying health indicators, such as obesity. Understanding the association between specific features of the built environment and obesity prevalence can lead to structural changes that could encourage physical activity and decreases in obesity prevalence.