时—空扫描统计量在传染病早期预警中的应用研究

发布时间：2018-03-08 22:27

本文选题：传染病监测　切入点：时空聚集性　出处：《四川大学》2007年博士论文　论文类型：学位论文

【摘要】： 目的传染病聚集性的早期发现是有效控制传染病爆发与流行、降低传染病发病率和死亡率的重要措施。但目前我国用于传染病聚集性探测的方法仅限于单纯的时间聚集性探测方法，不能充分利用数据信息，影响了聚集性探测的及时性和有效性。本研究旨在探讨时空扫描统计量这一先进的统计技术在传染病聚集性探测以及早期预警中的应用价值。方法模拟实时监测系统，采用时空扫描统计量建立模型，对我国2005年麻疹病例数据进行逐日前瞻性分析，并分别以省、区县和乡镇街道作为基本地理单位建立预警模型，在此基础上，探讨了探测结果三维可视化实现的途径和方法。结果全国省级分析结果显示：前瞻性时-空扫描统计量可提供确切的预警信息，对麻疹聚集性进行时间和空间定位；上海市、江苏省和浙江省的区县级分析结果显示：前瞻性时-空扫描统计量能及时探测到开始于某省局部县区和三省交界区县的麻疹聚集性。成都市乡镇街道级分析结果显示：在缺乏人口数据的情况下，前瞻性时-空重排扫描统计量仍能有效进行预警，且较前瞻性时间扫描统计量能更早探测到麻疹聚集性。结论时空扫描统计量能够充分挖掘和利用传染病监测数据中的时间和空间信息，提高聚集性探测的及时性和有效性，能够在各级行政区划早期探测到传染病在时空两个维度的聚集性。利用前瞻性时-空扫描统计量模型结合地理信息软件建立的早期预警系统，能够有效地做出早期预警。时空扫描统计量在实际工作中具有潜在的重要应用价值。
[Abstract]:Objective the early detection of infectious disease aggregation is to effectively control the outbreak and prevalence of infectious diseases. The important measures to reduce the morbidity and mortality of infectious diseases. However, at present, the methods used to detect the aggregation of infectious diseases in our country are limited to the simple method of detecting the aggregation of infectious diseases, and the data information is not fully utilized. The purpose of this study is to explore the application value of spatio-temporal scanning statistics, an advanced statistical technique, in the detection and early warning of infectious diseases. Methods the real-time surveillance system was simulated, and the model was established by space-time scanning statistics. The measles case data in 2005 in China were analyzed day by day, and the early warning model was established with provincial, district, county and township streets as the basic geographical units respectively. On this basis, the approaches and methods to realize 3D visualization of detection results are discussed. Results the results of provincial level analysis showed that the prospective time-space scanning statistics could provide accurate early warning information and locate the aggregation of measles in time and space. The results of district and county level analysis in Jiangsu province and Zhejiang province show that the prospective time-space scanning statistics can detect the measles aggregation in some counties and the borderline counties of three provinces in time. Analysis of Township and Township Street level in Chengdu. Shows: in the absence of population data, Prospective time-space rearrangement scan statistics can still be used for early warning, and measles aggregation can be detected earlier than prospective time-scanning statistics. Conclusion Spatio-temporal scanning statistics can fully mine and utilize the temporal and spatial information of infectious disease surveillance data, and improve the timeliness and effectiveness of aggregation detection. Early detection of the aggregation of infectious diseases in two dimensions of time and space in administrative divisions at all levels. An early warning system based on prospective time-space scanning statistics model and geographic information software, Spatio-temporal scanning statistics have potential important application value in practical work.
【学位授予单位】：四川大学
【学位级别】：博士
【学位授予年份】：2007
【分类号】：R181.3

【引证文献】