人群流行病学的空间分布模型仿真研究

发布时间：2018-12-12 01:20

【摘要】：在流行病人群分布模型的研究中,人群流行病学在空间上呈现非周期、几何指数、非线性的特征,分布过程复杂度较高,很难运用单一模型控制。传统的模型仅以小周期模型为基础,对大规模疾病分布建模的过程,以多模型组合逼近完成,缺少约束过程,造成复杂度很高,建模效果不好。为了避免上述缺陷,提出基于约束分类优化算法的人群流行病学空间分布模型。根据约束分类优化相关原理,对流行病的空间分布进行分层约束,得到较为合理的人群流行病种群的分布情况,对流行病种群中的所有个体进行编号处理,通过迭代计算得到最优分布区域的计算结果个体。通过临近地区流行病分布比例,得到人群流行病学的空间位置分布情况。实验结果表明,利用改进算法进行人群流行病学空间分布,能够极大的提高分布建模的准确性,降低建模复杂度。
[Abstract]:In the study of epidemic population distribution model, population epidemiology presents aperiodic, geometric exponent, nonlinear characteristics in space, the complexity of distribution process is high, it is difficult to use a single model to control. The traditional model is only based on the small-period model, and the modeling process of large-scale disease distribution is completed by multi-model combination approach, which is lack of constraint process, resulting in high complexity and poor modeling effect. In order to avoid these defects, a population epidemiology spatial distribution model based on constrained classification optimization algorithm is proposed. According to the principle of constraint classification optimization, the spatial distribution of epidemic is stratified, and the distribution of population epidemic population is obtained, and all individuals in epidemic population are numbered. The individual results of the optimal distribution region are obtained by iterative calculation. The spatial distribution of population epidemiology was obtained by epidemic distribution in adjacent areas. The experimental results show that the improved algorithm can greatly improve the accuracy of distribution modeling and reduce the complexity of modeling.
【作者单位】：广西医科大学信息管理与信息系统(医学)系;
【基金】：信息中心实验动物信息平台电子物理网络支撑体系建设应用研究(2060503科技条件专项) 2012年自治区科技基础条件平台建设财政补助项目
【分类号】：R181;TP391.9

【相似文献】