Abstract:

Objective To evaluate the simulation effect of Schistosoma japonicum Cellular Automata (SjCA) model by using the multi-dimensional evaluation system. Methods Several selected indicators, including the Youden index, sensitivity, specificity, accuracy rate and prevalence distance were applied to assess the validity of the SjCA model. The effectiveness of each indicator for model evaluation was compared. Results Different indicators in the multi-dimensional evaluation system generated assorted combination of optimized parameters for the SjCA model. The best model according to prevalence distance (99.97%) did not guarantee the ultimate fit when using accuracy rate (79.70%) as the evaluation indicator. Conclusion The multi-dimensional evaluation system is helpful for accurately assessing the simulation results of the discrete stochastic models such as SjCA, and is promising to be universalized to evaluate other similar epidemic dynamics models.

Key words: Schistosomiasis japonica, Transmission dynamics, Cellular automata model, Multi-dimensional indicator, Simulation effect

摘要：

目的 采用多维度指标对基于细胞自动机的日本血吸虫病传播动力学模型（SjCA）的模拟效果进行评价。方法 用约登指数、灵敏度、特异度、准确率、流行率距离等指标对SjCA模型的模拟输出进行分析，判断其对模拟有效性评价的效果。结果 根据不同评价指标，在对模型进行优化时，最佳参数组合不同。在SjCA模型中，群体流行率距离最大（99.97%）的参数组合，其个体准确率并非最大（79.70%）。结论 多维度指标评价体系有利于深入分析以细胞自动机模型为代表的离散随机模型的模拟效果，并有望被应用于其他流行病动力学模型的评价。

关键词: 日本血吸虫病, 传播动力学, 细胞自动机模型, 多维度指标, 模拟效果