全球78个国家和地区流行血吸虫病，在我国被列为重大传染病。钉螺是日本血吸虫病的唯一中间宿主，孳生地面积是指示感染性钉螺密度的首要指标，其有效识别受到广泛关注。以往的研究专注在应用遥感技术探测钉螺孳生地，错分率较高，至今尚无突破性研究进展。本项目将以空间信息技术为核心，在解决多种来源数据的整合与空间数据标准化的基础上，将机器学习方法与生态位理论结合，建立适用于高效识别钉螺孳生地的智能模型技术，并开发集成不同模型的开源平台系统，从全局和局部的空间角度全面比较与评价模型的效果。申请者的前期工作初步证明该研究思路的可行性和智能模型的优越性，其顺利实施将阐明钉螺孳生地分布的规律，明确其主要驱动因素，确定风险分布与环境变化背景下传播风险的变动趋势，为实现钉螺孳生地的现代化智能监测奠定必需的技术基础，为血吸虫病监测网络的完善提供新的方法与手段，也为其它媒介传播与环境相关疾病的智能化监测提供研究思路。

Schistosomiasis is endemic in 78 countries and regions in the world, and was listed as a major infectious disease in China. Oncomelania hupensis is the sole intermediate host of Schistosoma japonicum. The area of snail habitats is the leading indicator of infected snail density, so how to effectively identify snail habitats has attracted much attention. In the past ten years, most of the work was focused on exploring the technique of remote sensing to detect snail habitats, but its misclassification error is high and so far there is no significant breakthrough. On the basis of spatial information techniques, this study will first solve the issues of integrating and standardizing the spatial data from multiple sources, and then concentrate on studying the intelligent approach of effectively identifying snail habitats by applying the machine learning algorithm to the theory of ecological niche. Also, an open source platform system of merging different ecological niche models will be developed for comprehensively comparing and evaluating the model validity and accuracy from the global and local perspectives. Our preliminary work has proved the feasibility of this research idea and the advantages of intelligent models compared to the technique of remote sensing. The obtained results will help to illuminate the distribution patterns of snail habitats, discover the major driving factors, and determine its risk pattern and the changing trend of transmission risk under the background of environmental change. It is the necessary technical foundation for intelligently monitoring snail habitats, offers the new methods and means for improving the monitoring network of schistosomiasis, and provides the research ideas for intelligent monitoring of other vector-borne diseases and environment-related diseases.