Simulation of the Spread of Epidemic Disease Using Persistent Surveillance Data
This paper proposes a novel data-mining framework to simulate the spread of epidemic diseases using persistent surveillance data. The framework is formulated by merging the persistent surveillance data about epidemics, geographic information and the dynamics of disease into a heat transfer model according to the theory of statistical mechanics . In the implementation of this framework, geographic conditions are used to define the heat transfer media, which is featured by heat conductivity and thermal capacitance; persistent surveillance data about epidemic disease is used as the initial conditions; susceptible-infected- recovered (SIR) model, one of the most fundamental dynamic models about epidemic disease is employed as Neumann boundary conditions. As a result, the spread of epidemics can be simulated by solving the corresponding transient heat-transfer problem. Using COMSOL Multiphysics as the major platform, the framework is assessed by simulating the spread of a flu epidemic at a sample site in the Minneapolis (Minnesota, USA) region.