Network-based Analysis of Access Disparities in Medical Services

We delineated one-hour drive-time catchment areas for medical services using the road network and speed limits, to examine potential disparities in healthcare accessibility across different social groups.

Anisotropic Spatial Dependencies in Bike-Sharing

We propose an interpretable deep learning approach to uncover spatial relationships embedded in bike-sharing activities. We found that spatial influence does not strictly follow geographic proximity and is anisotropic. Non-member users exhibit weaker spatial dependencies in their bike usage behavior, resulting in lower short-term predictability compared to member users.

Paper: A Spatially-Adapted SHAP Approach for Interpreting Deep Bike Usage Learning and Prediction

Environmental Perceptions and Traffic Crashes

This study investigates the relationship between the perceived built environment and traffic crashes by employing computer vision techniques and machine learning methods. The findings offer insights into the mechanistic intersection of environmental perceptions and traffic safety.

Paper: Perceived built environment and non-motorist crashes: An exploration with street view imagery

In the Media: > Emotional perceptions can be a tool for predicting street safety

> Study: Boring Roads End Up With More Injuries For People Outside Cars, and beautiful roads report fewer

Spatiotemporal Modeling of Traffic Crash Risk

Traditional crash risk assessment methods, especially those aiming for cluster detection, are subject to the modified temporal unit problem, referring to the temporal effects (i.e., aggregation, segmentation, and boundary) in cluster detection. This study employs an emerging hot spot detection approach—the prospective space-time scan statistic (STSS) method—to assess crash risk at a refined network scale and across multiple years in Hartford, Connecticut. By identifying the spatial and temporal clusters of crash risk, the study provides evidence to support neighborhood-level road safety management strategies in areas characterized by elevated crash risk.

Paper: Assessing network-based traffic crash risk using prospective space-time scan statistic method