High-Dimensional Contact Network Epidemiology

Area of Science:

• Epidemiology and Network Science
• Mathematical Modeling of Infectious Diseases

Background:

• Traditional equation-based models have limitations in capturing complex disease transmission dynamics.
• Contact network models provide a more realistic framework for understanding disease spread.
• Recent advancements explore network dynamics and adaptive behaviors influencing transmission.

Purpose of the Study:

• To model disease spread on contact networks using bond percolation.
• To investigate the impact of edge weights derived from various independent variables on disease transmission.
• To compare the performance of contact network models against equation-based models for disease spread estimation.

Main Methods:

• Utilized bond percolation on weighted contact graphs to simulate disease spread.
• Edge weights were calculated as the product of probabilities of independent events involving multiple variables.
• Experiments included flight passenger data (US) and household contact data (Kenya, 2012).

Main Results:

• Contact network models demonstrated superior performance in estimating the spread of the 1918 Influenza virus compared to equation-based models.
• Edge weight calculations incorporating multiple variables provided nuanced insights into transmission dynamics.
• Exploration of network dynamics and adaptive features revealed key factors influencing disease propagation.

Conclusions:

• Contact network models, particularly those employing bond percolation with variable-weighted edges, offer a more accurate approach to epidemiological modeling.
• The methodology effectively captures the complexity of disease spread, outperforming traditional methods.
• Further research into adaptive network dynamics can enhance predictive capabilities for infectious disease outbreaks.