Presenters:
Shaokun Lyu
Qiluo Li
Haiyang Li
University of Southern California
“Environmental and socio-demographic factors play critical roles in shaping public health outcomes. This project applies a tri-environmental GeoAI framework to examine the spatial relationships between environmental exposures, socio-demographic characteristics, and five key health indicators: the prevalence of high blood pressure, coronary heart disease, diabetes, high cholesterol, and obesity. Using fine-grained data from Los Angeles County, we incorporate variables such as housing density, ethnic composition, green space density, water body density, and other built and natural environmental factors to understand their combined impact on health disparities.
Leveraging ArcGIS Pro for geospatial analysis and machine learning methods, including global and local random forest regression models, this study reveals spatial heterogeneity in health outcomes and identifies areas with heightened vulnerability. The project delivers tract-level maps and an analytical framework that can guide policy interventions aimed at addressing health inequities. By integrating environmental, social, and built-environment data, this research offers a replicable and transferable methodology that highlights the interplay of diverse factors influencing public health. The insights gained provide a robust foundation for informed decision-making and equitable policy development, underscoring the potential of GeoAI in tackling complex health challenges.”
