A0575
Title: Unraveling hybrid diffusion structures in epidemics: A Bayesian spatiotemporal modeling framework
Authors: Tzai-Hung Wen - National Taiwan University (Taiwan) [presenting]
Abstract: Understanding the spatial structure of disease transmission is essential for analyzing the diffusion dynamics of epidemics across time and space. Epidemic spread commonly follows expansion, relocation, or hybrid diffusion patterns, each shaped by distinct transmission mechanisms. Traditional spatial statistical models, which rely on adjacency or mobility-based assumptions, often fail to capture the complex and heterogeneous nature of these processes, particularly when hybrid patterns emerge. To address this limitation, a Bayesian framework is introduced using a conditional autoregressive adaptive model to estimate the spatial structure of dengue transmission in Tainan City, Taiwan. This approach enables the data-driven construction of spatial weight matrices that capture both localized expansion and broader relocation dynamics. The model successfully identifies spatial heterogeneity and interregional dependencies in transmission pathways, offering a more realistic representation of disease spread. By capturing these nuanced diffusion structures, methodological approaches are advanced to spatial epidemiology, and the interpretability of spatial interactions is enhanced in disease dynamics. The findings provide valuable insights for designing context-specific public health interventions and contribute to the development of more effective strategies for epidemic prevention and control.
