Format

Send to

Choose Destination
Sci Rep. 2019 Apr 18;9(1):6229. doi: 10.1038/s41598-019-42036-w.

Measuring social, environmental and health inequalities using deep learning and street imagery.

Suel E1,2,3, Polak JW4,5, Bennett JE6,7, Ezzati M6,7,8.

Author information

1
School of Public Health, Imperial College London, London, UK. esra.suel@imperial.ac.uk.
2
MRC-PHE Centre for Environment and Health, Imperial College London, London, UK. esra.suel@imperial.ac.uk.
3
Urban Systems Laboratory, Imperial College London, London, SW7 2AZ, United Kingdom. esra.suel@imperial.ac.uk.
4
Urban Systems Laboratory, Imperial College London, London, SW7 2AZ, United Kingdom.
5
Centre for Transport Studies, Department of Civil and Environmental Engineering, Imperial College London, London, SW7 2AZ, United Kingdom.
6
School of Public Health, Imperial College London, London, UK.
7
MRC-PHE Centre for Environment and Health, Imperial College London, London, UK.
8
WHO Collaborating Centre on NCD Surveillance and Epidemiology, Imperial College London, London, UK.

Abstract

Cities are home to an increasing majority of the world's population. Currently, it is difficult to track social, economic, environmental and health outcomes in cities with high spatial and temporal resolution, needed to evaluate policies regarding urban inequalities. We applied a deep learning approach to street images for measuring spatial distributions of income, education, unemployment, housing, living environment, health and crime. Our model predicts different outcomes directly from raw images without extracting intermediate user-defined features. To evaluate the performance of the approach, we first trained neural networks on a subset of images from London using ground truth data at high spatial resolution from official statistics. We then compared how trained networks separated the best-off from worst-off deciles for different outcomes in images not used in training. The best performance was achieved for quality of the living environment and mean income. Allocation was least successful for crime and self-reported health (but not objectively measured health). We also evaluated how networks trained in London predict outcomes three other major cities in the UK: Birmingham, Manchester, and Leeds. The transferability analysis showed that networks trained in London, fine-tuned with only 1% of images in other cities, achieved performances similar to ones from trained on data from target cities themselves. Our findings demonstrate that street imagery has the potential complement traditional survey-based and administrative data sources for high-resolution urban surveillance to measure inequalities and monitor the impacts of policies that aim to address them.

Supplemental Content

Full text links

Icon for Nature Publishing Group Icon for PubMed Central
Loading ...
Support Center