湖南省肾综合征出血热的流行特征及Elman神经网络预测

发布时间：2018-10-16 22:06

【摘要】： 目的：阐述2000年以来肾综合征出血热的流行特征,探讨肾综合征出血热流行的影响因素,进行肾综合征出血热宿主的流行病学调查,了解宿主的分布及带毒率情况,建立对肾综合征出血热发病人数的Elman神经网络模型,对预测结果进行为更好的进行肾综合征出血热的防治提供参考。方法：对湖南省2000年以来的法定肾综合征出血热疫情报告数据资料及宿主监测资料,用Excel2003和SPSS13.0统计软件进行统计分析。应用MATLAB软件对肾综合征出血热的发病人数进行预测,建立Elman预测模型。结果：2000年-2008年以来全省共发生肾综合征出血热病例11036例,平均发病率为1.88／10万,发病地区大致没有变化,发病人数男性多于女性,发病年龄以30-50的青壮年为主。发病的职业分布上,农民的发病人数最多,占74.35%,其次是工人,占6.66%。不同鼠种在室内外的分布上存在差异(χ2=295.778,p=0.000),褐家鼠为室内的优势鼠种而黑线姬鼠为室外的优势鼠种。不同鼠种在室内外密度上也存在差异(χ2=92.513,p=0.000),但不同鼠种的带毒率之间总体上不存在差异(χ2=1.678,p=0.795),暗示不同鼠种之间可能存在交叉传播,不同鼠种以黑线姬鼠的带毒指数最高,褐家鼠居其次,这两种鼠可能是肾综合征出血热传播的主要宿主。运用Elman神经网络对肾综合征出血热(Hemorrhagic Fever with Renal Syndrome, HFRS)的发病人数得到实际发病人数为47人,预测人数为56人,修正后预测人数为52人,预测人数与实际人数基本一致。结论：湖南省HFRS发病率前4年来持续下降,发病仍以男性为主,发病年龄仍以青壮年为主,发病职业主要以农民和工人为主。室内鼠种以褐家鼠为优势鼠种,室外鼠种以黑线姬鼠为优势鼠种。预测和修正后误差与实际人数基本一致,采用Elman神经网络对HFRS发病人数进行预测的方法是可行的。
[Abstract]:Objective: to describe the epidemic characteristics of hemorrhagic fever with renal syndrome (HFRS) since 2000, to explore the influencing factors of hemorrhagic fever with renal syndrome (HFRS), to investigate the host epidemiology of hemorrhagic fever with renal syndrome (HFRS), and to understand the distribution of the host and the prevalence of hemorrhagic fever with renal syndrome (HFRS). A Elman neural network model for the number of hemorrhagic fever with renal syndrome (HFRS) was established to provide a reference for better prevention and treatment of hemorrhagic fever with renal syndrome (HFRS). Methods: the reported data and host surveillance data of hemorrhagic fever with renal syndrome (HFRS) in Hunan Province since 2000 were analyzed by Excel2003 and SPSS13.0 software. MATLAB software was used to predict the incidence of hemorrhagic fever with renal syndrome (HFRS), and a Elman predictive model was established. Results: from 2000 to 2008, there were 11036 cases of hemorrhagic fever with renal syndrome in the whole province, the average incidence rate was 1.88 / 100 000. The number of males was more than that of females, and the age of onset was 30-50 years old. In the occupational distribution of the disease, the number of farmers was the most, accounting for 74.35, followed by workers, accounting for 6.66. The distribution of Rattus norvegicus and Apodemus agrarius were different in indoor and outdoor (蠂 2 + 295.778P0. 000), and Rattus norvegicus was the dominant species in indoor and Apodemus agrarius outdoors. There were also differences in indoor and outdoor density among different rodent species (蠂 ~ 2 ~ (2 +) 92.513), but there was no difference among different mouse species (蠂 ~ (2) (蠂 ~ (2) (蠂 ~ (2) (蠂 ~ (2) 1.678 (P _ (0.795), suggesting that there might be cross-propagation among different mouse species, and the virus carrying index of Apodemus agrarius was the highest in different mouse species. Rattus norvegicus is next, these two kinds of mice may be the main host of hemorrhagic fever with renal syndrome. The actual incidence of hemorrhagic fever with renal syndrome (Hemorrhagic Fever with Renal Syndrome, HFRS) by Elman neural network was 47, the predicted number was 56, and the revised number was 52. The predicted number was basically consistent with the actual number. Conclusion: the incidence of HFRS in Hunan Province has been decreasing continuously in the past 4 years. The incidence of HFRS is still mainly male, the age of onset is still young and adult, and the occupation is mainly farmers and workers. Rattus norvegicus was the dominant rodent in indoor species and Apodemus agrarius was dominant in outdoor rodent species. The error after prediction and correction is basically consistent with the actual number of people. It is feasible to use Elman neural network to predict the number of HFRS patients.
【学位授予单位】：中南大学
【学位级别】：硕士
【学位授予年份】：2010
【分类号】：R512.8;R181.3

【相似文献】