Infectious Disease Spread Prediction Models and Consideration
Sakumura, Takenori - Kyushu Institute of Technology
Matsuguma, Kazuhiro - Kyushu Institute of Technology
Hirose, Hideo - Kyushu Institute of Technology
For infectious disease spread prediction models, pandemic simulations have been dealt with as a kind of simulation by scenario. However, when a pandemic occurs, predicting the future using the observed data becomes crucial. The methodology that assumes the model structure and estimates the model parameters using the observed data, which is called the assimilation or the gray box, is also necessary in the pandemic analysis. In this paper, we propose a method to estimate such parameters, called the BBS (best-backward solution) method, and discuss the prediction results for observed real cases such as the SARS and the FMD (foot-and-mouth disease). We compare the results with those using the truncated model. We have found that the SIR model provides the worst case predictions even in early stages of pandemics contrary to the truncated model.
The Information Processing Society of Japan, Transactions on Mathematical Modeling and its Applications (TOM), Volume 4, Number 3, Pages 102–109, July 2011.