基于遥感和智能体的鄱阳湖区血吸虫病监测和模拟研究

发布时间：2018-02-23 00:04

本文关键词： 血吸虫病鄱阳湖洲滩钉螺遥感决策树特征提取 BP神经网络智能体分布式计算 Swarm Java3D VRML　出处：《中国科学院研究生院（遥感应用研究所）》2006年博士论文　论文类型：学位论文

【摘要】：随着空间信息技术的更新和发展，其在公共卫生领域的应用也更加深入和广泛。遥感能够快速、全面地探索、发现、监测与人类健康相关的环境因素，地理信息系统为传染病相关流行因素数据的管理和分析提供了快速、先进、可靠的技术手段，结合多维信息可视化、计算机模拟等数据分析与表达方法不仅能动态分析传染病的时间与空间分布特征，而且可以使我们从全新的角度和方式来研究和认识传染病，从而为传染病时空传播和流行规律的研究提供新的思路和方法。血吸虫病是一种具有地方性和自然疫源性的人畜共患传染病，在我国存在历史悠久，分布面积广泛，传播过程又涉及众多的自然因素和社会因素，，从而给血吸虫病防制工作带来了极大的困难。近几年来，由于西部大开发、三峡移民等工程的实施，我国的自然、社会和经济等因素变化较大，血吸虫病疫情回升显著，表现为血吸虫病患病人数增多，急性感染人数呈上升趋势，局部地区钉螺扩散明显，感染性钉螺分布范围逐渐扩大，部分已经达到传播控制标准和传播阻断标准的地区疫情严重回升，出现向城市蔓延的趋势等，对人民健康、经济发展和社会进步构成威胁，血防工作形势严峻。我国血吸虫病流行区具有严格的地方性，它与钉螺的分布基本一致。可分为水网、湖沼和山丘三种不同类型。流行区中，湖沼型钉螺面积占全国钉螺总面积3400km~2的95％以上，其中，江西省鄱阳湖周围的11个县为血吸虫病流行区。本文基于空间信息技术研究鄱阳湖区洲滩环境下血吸虫病相关因素的监测方法及其时空传播规律，旨在为我国湖区GIS血吸虫病预警系统的建设提供方法和技术的贡献，从而为血吸虫病防制提供辅助决策和为应急反应奠定基础。本文主要开展了以下四部分的研究工作。 (1) 基于TM遥感影像的钉螺可疑孳生地的识别研究。针对鄱阳湖区钉螺分布面积广泛的特点，本文基于流行病学现场调查的方法，在2005年现场考察鄱阳湖区血吸虫病流行地区，同时测量和收集疫区洲滩上的植被数据和钉螺数据。结合历史文献资料以及合作单位提供的数据，分析了鄱阳湖区洲滩和有螺洲滩的分布状况，进而分析了鄱阳湖区植被分布状况及其与钉螺分布的关系。采用同一时相的覆盖鄱阳湖区的TM遥感影像，选取合适的研究区，仔细分析研究区地貌，典型地物的光谱特征，地面考察收集样本点的光谱特征，形成用于分类的专家经验性知识，然后采用基于知识模型的专家决策分类方法分类识别钉螺可疑孳生地，分类精度较已做过的研究有提高，总体精度为88.94％。将分类结果与鄱阳湖区钉螺分布地图集进行了对比分析，并统计分析了2005年研究区内钉螺孳生面积情况，结果表明整个研究区面积为1567.720km~2，其中适合钉螺孳生的面积
[Abstract]:With the updating and development of spatial information technology, its application in the field of public health has become deeper and wider. Remote sensing can rapidly and comprehensively explore, detect and monitor environmental factors related to human health, Geographic Information system (GIS) provides rapid, advanced and reliable technical means for the management and analysis of epidemic factors related to infectious diseases. The methods of data analysis and expression, such as computer simulation, can not only dynamically analyze the temporal and spatial distribution of infectious diseases, but also enable us to study and understand infectious diseases from a completely new perspective and way. Therefore, it provides new ideas and methods for the study of the temporal and spatial spread and epidemic law of infectious diseases. Schistosomiasis is a zoonotic infectious disease with local and natural epidemic origin. Schistosomiasis has a long history and is widely distributed in China. The transmission process involves many natural and social factors. This has brought great difficulties to the prevention and control of schistosomiasis. In recent years, due to the great development of the western region and the implementation of the three Gorges resettlement project, the natural, social and economic factors in our country have changed greatly, and the epidemic situation of schistosomiasis has picked up significantly. The results showed that the number of schistosomiasis cases increased and the number of acute infections increased. The distribution of infected snails was gradually expanded in some areas, the spread of Oncomelania hupensis was obvious in some areas. The epidemic situation in some areas which have reached the standard of communication control and blocking has risen seriously, and the trend of spreading to cities has emerged, which poses a threat to the health of the people, economic development and social progress, and the situation of blood control work is grim. The schistosomiasis endemic area in China is strictly endemic, which is basically consistent with the distribution of snail. It can be divided into three types: water net, lake and marsh, and hilly. In the endemic area, the area of lake-marsh snail accounts for more than 95% of the total snail area of 3400kmm-2 in China. The 11 counties around Poyang Lake in Jiangxi Province are endemic areas of schistosomiasis. Based on spatial information technology, this paper studies the monitoring method of schistosomiasis related factors and its space-time transmission law in the beach environment of Poyang Lake region. In order to provide methods and techniques for the construction of GIS schistosomiasis warning system in the lake area of China, and to provide a basis for decision making and emergency response to schistosomiasis prevention and control, four parts of research work are carried out in this paper. 1) Identification of suspected breeding ground of Oncomelania hupensis based on TM remote sensing image. In view of the widespread distribution of Oncomelania hupensis in Poyang Lake region, based on the method of epidemiological field investigation, the endemic area of schistosomiasis in Poyang Lake region was investigated in 2005. At the same time, the vegetation data and snail data were measured and collected on the beaches in the epidemic area. Based on the historical literature and the data provided by the cooperating units, the distribution of the beaches and the beaches with Luozhou in the Poyang Lake region were analyzed. The distribution of vegetation and the relationship between vegetation distribution and snail distribution in Poyang Lake region are analyzed. The TM remote sensing image covering Poyang Lake region with the same time phase is used to select the suitable research area, and the spectral characteristics of the landforms and typical features of the study area are carefully analyzed. The spectral characteristics of sample points were collected on the ground to form expert empirical knowledge for classification. Then the expert decision classification method based on knowledge model was used to classify and identify the suspected breeding areas of Oncomelania hupensis. The classification accuracy was improved compared with the previous research. The total accuracy is 88.94. The classification results are compared with the distribution atlas of Oncomelania hupensis in Poyang Lake region, and the breeding area of Oncomelania hupensis in 2005 is statistically analyzed. The results show that the area of the whole study area is 1 567.720 km / m ~ 2, and the area suitable for snail breeding is suitable for snail breeding.
【学位授予单位】：中国科学院研究生院（遥感应用研究所）
【学位级别】：博士
【学位授予年份】：2006
【分类号】：R184

【引证文献】