Extending Participatory Sensing to Personal Exposure Using Microscopic Land Use Regression Models

Area of Science:

• Environmental Health Sciences
• Epidemiology
• Data Science

Background:

• Personal exposure to environmental factors varies significantly due to individual behavior and characteristics.
• Participatory sensing offers a method to capture spatial and temporal environmental variability.
• Bridging participatory sensing data with health research is crucial for future epidemiological studies.

Purpose of the Study:

• To present a novel methodology, instantaneous microscopic land-use regression (µLUR), to connect participatory sensing with health research.
• To develop a data workflow for applying µLUR models to mobile populations.
• To demonstrate the application of µLUR for modeling Black Carbon exposure.

Main Methods:

• Utilized data science techniques to extract activity-specific and route-sensitive spatiotemporal variability.
• Developed a data workflow for preparing and applying µLUR models.
• Integrated instantaneous noise assessments for real-time traffic data.
• Created activity-specific models for bicycle, in-vehicle, and indoor micro-environments.

Main Results:

• Successfully modeled instantaneous personal exposure to traffic-related Black Carbon.
• Achieved a correlation of 0.65 in independent external validation.
• Demonstrated the applicability of µLUR across different micro-environments.

Conclusions:

• Instantaneous microscopic land-use regression (µLUR) effectively models personal exposure variability.
• The proposed data workflow facilitates the application of µLUR in diverse epidemiological contexts.
• µLUR models can be applied to simulated behaviors for advanced health and policy research.