The use of remote sensing for predictive modeling of schistosomiasis in China

Abstract

The development of predictive models of the spatial distribution of schistosomiasis are hampered by the existence of different regional subspecies of the Oncomelania hupensis snail that serve as intermediate hosts for the disease in China. The habitats associated with these different subspecies vary considerably, with mountainous habitats in the west and floodplain habitats in the east. Despite these challenges, continuing environmental change resulting from the construction o/the Three Gorges Dam and global warming that threaten to increase snail habitat, as well as limited public health resources, require the ability to accurately map and prioritize areas at risk for schistosomiasis. This paper describes a series of ongoing studies that rely on remotely sensed data to predict schistosomiasis risk in two regions of China. The first study is a classification of tandsat TM imagery to identify snail habitats in mountainous regions of Sichuan Province. The accuracy of this classification was assessed in an independent field study, which revealed that seasonal flooding may have contributed to misclassification, and that the incorporation of soil maps may greatly improve classification accuracy. A second study presents the use of tandsat TM and water level data to understand seasonal differences in Oncomelania hupensis habitat in the lower Yangtze River region.