Obesity is associated with numerous health outcomes including diabetes, heart disease, and cancer. Over 30% of adults and 17% of children in the US are obese, with lack of physical activity (PA) and constraints in the built environment (BE) common culprits. As such, the US Guide to Community Preventive Services currently recommends the following BE interventions to increase PA and reduce obesity: (1) community and street-scale urban design and land use policies; (2) creation of, or enhanced access to places for physical activity; and (3) transportation policies and practices. The proposed project will use an archive of 23,000 publically-available, online, outdoor webcams and crowdsourcing to develop reliable and valid tools to improve the capture of global PA patterns and BE characteristics.
The specific aims are to: 1) Develop and test the reliability of using publically-available, outdoor webcams to enumerate BE characteristics and PA patterns across thousands of global outdoor environments; and 2) Develop and test the reliability and validity of using crowdsourcing to enumerate BE characteristics and PA patterns. (13) The aims will be accomplished by using the Archive of Many Outdoor Scenes (AMOS) that has collected over 325 million images of outdoor environments from more than 23,000 webcam since 2006. Every identified, outdoor, publically-available online webcam is added to the AMOS dataset with an image taken and archived every 30 minutes from the webcam thereafter. Scenes include street intersections, plazas, and parks. We will use crowdsourcing to annotate each webcam scene. The crowdsourcing platform is Amazon Mechanical Turk (AMT), an Internet marketplace allowing people to be paid to complete small, computer-based tasks. The individual tasks are posted to the AMT website. We will use AMT tasks to annotate each AMOS webcam scene by identifying BE characteristics (established from the Active Neighborhood Checklist and Public Open Space Tool) and patterns of PA (established from SOPARC). This study is significant in its ability to increase by orders of magnitude the amount and quality of global recorded measurements of PA patterns and associated BE characteristics. This project is innovative by: 1) combining two distinct disciplines to use public, outdoor webcams to evaluate PA patterns and BE characteristics; 2) utilizing crowdsourcing to provide simple PA counts and indication of BE characteristics; and 3) its ability to greatly improve the generalizability of public health surveillance. Our results will yield actionable knowledge regarding the use of webcams and crowdsourcing to assess PA and the BE.
This project aims to combine computer science and public health to use online, outdoor webcams and crowdsourcing to enumerate global physical activity patterns and built environment characteristics. Webcams and crowdsourcing will be evaluated for reliability and validity. Results will yield actionable knowledge regarding the use of webcams and crowdsourcing to assess PA patterns and BE characteristics.
| Adlakha, Deepti; Budd, Elizabeth L; Gernes, Rebecca et al. (2014) Use of emerging technologies to assess differences in outdoor physical activity in st. Louis, missouri. Front Public Health 2:41 |
| Kaczynski, Andrew T; Wilhelm Stanis, Sonja A; Hipp, J Aaron (2014) Point-of-decision prompts for increasing park-based physical activity: a crowdsource analysis. Prev Med 69:87-9 |
| Graham, Dan J; Hipp, J Aaron (2014) Emerging technologies to promote and evaluate physical activity: cutting-edge research and future directions. Front Public Health 2:66 |
| Hipp, J Aaron; Adlakha, Deepti; Eyler, Amy A et al. (2013) Emerging technologies: webcams and crowd-sourcing to identify active transportation. Am J Prev Med 44:96-7 |
