Prevention & lipidsCohort study
Deep learning analysis of Google Street View to assess residential built environment and cardiovascular risk in a Midwestern US retrospective cohort
- Retrospective cohort study
Authors' conclusion
Visible vertical greenspace and sidewalks, as discerned from street-level images using deep learning, demonstrated potential associations with cardiovascular risk.
Study context
- Population
- Residents of a Northeast Ohio cohort, built environment assessed by deep learning of Google Street View imagery2,083 with major adverse cardiovascular events (MACE)
- Participants
- 49,887 residents
- Exposure
- Residential built environment features from Google Street View (GSV) imagery and MACE risk
- Outcome
- MACE
- Follow-up
- Median 26.86 months
Main finding
As reported in the abstract
Hazard ratio (95% CI)
Higher tree-sky index
0.95(0.91–0.99)
Sidewalk presence
0.91(0.87–0.96)
Observational data: associations, not effects.
Original publication
Deep learning analysis of Google Street View to assess residential built environment and cardiovascular risk in a Midwestern US retrospective cohort. Eur J Prev Cardiol. Published 4 February 2025. doi:10.1093/eurjpc/zwaf038
Summarised from the published abstract; every number is checked against it