Imagining Built Environment Changes: Examining the Potential of Incorporating Generative Artificial Intelligence into a Citizen Science Program for Older Women
Time: -Topics: Digital Health, Physical Activity
Background: In recent years, generative artificial intelligence (AI)—which allows users to generate text and photo content through internet-based programs such as ChatGPT—has generated substantial interest and concern. However, little is currently known about the potential of AI tools to contribute to physical activity (PA) promotion interventions. To contribute to this nascent field, we are evaluating the acceptability and perceived utility of incorporating simple AI into the Our Voice (OV) citizen science method for older women.
Methods: To supplement the 4-step OV method—which encourages citizen scientists to leverage the data they collect about their local environments to activate community-level change—we are evaluating two iterations of an OV “Imagine” step with and without AI to determine whether this technology can benefit the intervention. Two groups receiving the OV intervention in the ongoing Women’s Health Initiative Strong & Healthy (WHISH) Encore study—which provides citizen science programs to promote PA in US older women—were randomized to receive one of these Imagine Steps. For the group receiving AI, researchers are utilizing the Generative Fill technology from Adobe Photoshop (2025) to enrich images that participants took during their walk audits to visualize changes in their communities. In the non-AI group, researchers and participants are utilizing traditional online search engines to find images representative of their proposed changes to see how they have been implemented previously. Surveys and group discussions about the Imagine step with each group are being used to assess its perceived utility and acceptability among these older women. Additionally, baseline and 6-mo. assessments about civic engagement and neighborhood satisfaction will be used to determine the extent to which each Imagine intervention encourages participants to advocate for change in their communities.
Discussion: Imagine step implementation and data collection are ongoing. This investigation with a sample of older women represents, to our knowledge, the first evaluation of the acceptability and utility of this image-editing AI technology for a PA-promoting citizen science program. The survey findings and qualitative feedback from group discussions will provide novel insights into the benefits and challenges of employing such AI technologies and could inform future interventions utilizing them to promote healthy aging.
Keywords: Technology, Built environmentMethods: To supplement the 4-step OV method—which encourages citizen scientists to leverage the data they collect about their local environments to activate community-level change—we are evaluating two iterations of an OV “Imagine” step with and without AI to determine whether this technology can benefit the intervention. Two groups receiving the OV intervention in the ongoing Women’s Health Initiative Strong & Healthy (WHISH) Encore study—which provides citizen science programs to promote PA in US older women—were randomized to receive one of these Imagine Steps. For the group receiving AI, researchers are utilizing the Generative Fill technology from Adobe Photoshop (2025) to enrich images that participants took during their walk audits to visualize changes in their communities. In the non-AI group, researchers and participants are utilizing traditional online search engines to find images representative of their proposed changes to see how they have been implemented previously. Surveys and group discussions about the Imagine step with each group are being used to assess its perceived utility and acceptability among these older women. Additionally, baseline and 6-mo. assessments about civic engagement and neighborhood satisfaction will be used to determine the extent to which each Imagine intervention encourages participants to advocate for change in their communities.
Discussion: Imagine step implementation and data collection are ongoing. This investigation with a sample of older women represents, to our knowledge, the first evaluation of the acceptability and utility of this image-editing AI technology for a PA-promoting citizen science program. The survey findings and qualitative feedback from group discussions will provide novel insights into the benefits and challenges of employing such AI technologies and could inform future interventions utilizing them to promote healthy aging.
Authors and Affliiates
Presenter: Arjan S. Walia, MS, Stanford University Department of Epidemiology and Population HealthCo-Author: Cesar G. Lopez, BS, Stanford University Department of Epidemiology and Population Health
Co-Author: Maria I. Campero, BS, Stanford University Department of Epidemiology and Population Health
Co-Author: Zakaria N. Doueiri, BS, Stanford Univesity Department of Epidemiology and Population Health
Co-Author: Sally F. Mackey, MS, Stanford Prevention Research Center
Co-Author: Marcia L. Stefanick, PhD, Stanford Prevention Research Center
Co-Author: Abby C. King, PhD, FSBM, Stanford University Department of Epidemiology and Population Health
Imagining Built Environment Changes: Examining the Potential of Incorporating Generative Artificial Intelligence into a Citizen Science Program for Older Women
Category
Scientific > Rapid Communication Poster
