Template-driven spatial-temporal outbreak simulation for outbreak detection evaluation

Area of Science:

• Epidemiology
• Computational Biology
• Public Health Informatics

Background:

• Evaluating outbreak detection algorithms is crucial for public health surveillance.
• Existing simulators may lack flexibility or generate aggregated data, limiting their utility.
• A need exists for adaptable tools that simulate realistic outbreak scenarios.

Purpose of the Study:

• To develop a non-disease specific, template-driven spatial-temporal simulator.
• To generate diverse outbreak patterns for evaluating outbreak detection algorithms.
• To provide a flexible and easy-to-implement simulation tool.

Main Methods:

• Developed a template-driven spatial-temporal outbreak simulator.
• Implemented three generation algorithms: deterministic, independent, and Poisson process.
• Utilized linear template functions for simulation examples.
• Generated individual case event times, not aggregated counts.

Main Results:

• The simulator can generate various temporal and spatial-temporal outbreak patterns.
• The tool is flexible, easy to implement, and parameter-driven.
• Simulations provide detailed case event times for granular analysis.
• Demonstrated utility with linear template function examples.

Conclusions:

• The Template-Driven Simulator is a valuable tool for assessing outbreak detection algorithms.
• Its flexibility and event-time data enhance the evaluation of surveillance systems.
• Facilitates the development and validation of public health response strategies.