發(fā)布時(shí)間：2018-03-05 21:08

  本文選題：達(dá)烏爾黃鼠疫源地　切入點(diǎn)：風(fēng)險(xiǎn)分級(jí)　出處：《中國疾病預(yù)防控制中心》2013年碩士論文　論文類型：學(xué)位論文

【摘要】：鼠疫是嚴(yán)重危害人類健康的烈性傳染病,其發(fā)生和流行不僅威脅公眾的生命,而且會(huì)對(duì)經(jīng)濟(jì)發(fā)展產(chǎn)生嚴(yán)重的影響。預(yù)測(cè)預(yù)警技術(shù)是以早期發(fā)現(xiàn)傳染病異常為目的,為控制疫情的發(fā)展和傳播贏得了寶貴時(shí)間的一個(gè)新興技術(shù)和方法,我國傳染病預(yù)測(cè)預(yù)警技術(shù)正處于起步階段,但已逐漸成為一個(gè)研究的熱點(diǎn)領(lǐng)域。 我國的鼠疫流行已久,且疫源地種類多,面積大,疫情常年發(fā)生,因此其預(yù)測(cè)預(yù)警技術(shù)的研究就顯得十分必要。達(dá)烏爾黃鼠疫源地在我國的起源較早,經(jīng)歷了東北鼠疫的兩次大流行,雖然近年來疫情明顯減弱,但是發(fā)生動(dòng)物及人間鼠疫流行的風(fēng)險(xiǎn)依然存在。 本文主要利用Matlab軟件,應(yīng)用回歸分析中最優(yōu)回歸子集法及時(shí)間序列分析中指數(shù)平滑的方法對(duì)達(dá)烏爾黃鼠疫源地及疫源地內(nèi)6個(gè)國家級(jí)監(jiān)測(cè)點(diǎn)扎魯特旗、科右中旗、正白旗烏寧巴圖、建平縣、鎮(zhèn)賚縣及哈爾濱市的監(jiān)測(cè)數(shù)據(jù)進(jìn)行風(fēng)險(xiǎn)分級(jí)及預(yù)測(cè)預(yù)警。文中主要利用達(dá)烏爾黃鼠疫源地內(nèi)蒙古整體數(shù)據(jù)建立總體回歸方程模型并進(jìn)行風(fēng)險(xiǎn)分級(jí),回歸方程考慮以下7項(xiàng)作為影響鼠疫流行的因素：達(dá)烏爾黃鼠密度、達(dá)烏爾黃鼠鼠體染蚤率、達(dá)烏爾黃鼠鼠體蚤指數(shù)、巢穴蚤染蚤率、巢穴蚤蚤指數(shù)、洞干蚤染蚤率及洞干蚤蚤指數(shù),結(jié)果顯示當(dāng)選取達(dá)烏爾黃鼠密度、達(dá)烏爾黃鼠鼠體染蚤率、達(dá)烏爾黃鼠鼠體蚤指數(shù)及巢穴蚤染蚤率這4項(xiàng)指標(biāo)和選取多于這4項(xiàng)指標(biāo)對(duì)預(yù)測(cè)鼠疫發(fā)生的影響基本相同。風(fēng)險(xiǎn)分級(jí)為將檢出鼠疫菌視為流行,取值為1,未檢出菌視為不流行,取值為0,對(duì)待判數(shù)據(jù)的預(yù)報(bào)方法是將風(fēng)險(xiǎn)分為三級(jí),若預(yù)報(bào)值y2/3,則將該疫點(diǎn)預(yù)報(bào)為流行；若預(yù)報(bào)值y1/3,則預(yù)報(bào)為不流行；若1/3y2/3,則預(yù)報(bào)為高風(fēng)險(xiǎn)地區(qū),風(fēng)險(xiǎn)分級(jí)后利用實(shí)際數(shù)據(jù)進(jìn)行擬合,當(dāng)y2/3時(shí)預(yù)報(bào)流行的符合率均為100%；當(dāng)y1/3,回歸因子選取≥4個(gè)時(shí)預(yù)報(bào)流行的符合率均為100%；當(dāng)1/3y2/3時(shí),回歸因子選取≥4個(gè)時(shí)預(yù)報(bào)流行的擬合率大約為50%。對(duì)內(nèi)蒙古整體數(shù)據(jù)利用指數(shù)平滑方法預(yù)測(cè)2012年達(dá)烏爾黃鼠疫源地動(dòng)物鼠疫不流行,6個(gè)國家級(jí)監(jiān)測(cè)點(diǎn)數(shù)據(jù)的驗(yàn)證預(yù)測(cè)也不流行,達(dá)烏爾黃鼠疫源地2012年實(shí)際情況未檢出鼠疫菌,檢出血凝陽性材料1份,預(yù)測(cè)和實(shí)際情況基本符合。
[Abstract]:Plague is a severe infectious disease that seriously endangers human health. Its occurrence and prevalence not only threaten the lives of the public, but also have a serious impact on economic development. In order to control the development and spread of epidemic situation, it is a new technology and method that has won precious time. The prediction and early warning technology of infectious disease in China is in its infancy, but it has gradually become a hot field of research. Yersinia pestis has been prevalent for a long time in our country, and there are many kinds of foci, the area is large, and the epidemic occurs all the year round, so it is necessary to study the prediction and early warning technology of Yersinia pestis in our country. After two pandemics of plague in Northeast China, although the epidemic situation has weakened obviously in recent years, the risk of plague epidemic in animals and humans still exists. Using Matlab software, the optimal regression subset method in regression analysis and the exponential smoothing method in time series analysis were applied to the Zhalutte Banner and the middle flag of the family right in the foci of the Daour yellow rat and the 6 national monitoring points in the foci. The monitoring data of Zhengbai Banner, Wuning Gbatu, Jianping County, Zhenlai County and Harbin City were used to classify the risk and predict the risk. In this paper, the overall regression equation model was established and the risk classification was carried out by using the overall data of Daour yellow rat foci in Inner Mongolia. The regression equation considered the following seven factors as the factors influencing the plague epidemic: the density of Daour yellow rat, the flea rate of Daour rat, the flea index of Daour rat, the flea rate of nest flea and the flea index of nest flea. The results showed that when the density of Daour yellow rat was selected, the flea rate of Daour was determined. The body flea index of Daour rat and the flea infection rate of nest flea were basically the same. The risk classification was that the detected Yersinia pestis was regarded as epidemic, the value was 1, and the undetected bacteria was not epidemic. If the forecast value is zero, the forecast method for the judgment data is to divide the risk into three levels. If the forecast value is y2 / 3, the epidemic spot will be predicted as epidemic; if the predicted value is y1 / 3, the forecast will be non-epidemic; if the forecast value is y1 / 3 / 3, then the forecast will be a high-risk area. After risk classification, the coincidence rate of forecasting epidemic was 100 when y2 / 3:00, when y1 / 3, when regression factor 鈮，

本文編號(hào)：1571894