deleo-lab/schisto-vegetation-drone

Semantic segmentation of aquatic vegetation associated with schistosomiasis in Senegal Africa # Schisto-vegetation-drone CNNs, U-Net model, Keras, TensorFlow backened, Semantic segmentation ## Introduction Schistosomiasis is a debilitating parasitic disease of poverty that affects more than 200 million people worldwide, mostly in sub-Saharan Africa that has been associated with the construction of dams and water infrastructure in tropical areas. Changes in hydrology and salinity associated with water infrastructure create conditions favorable to the aquatic vegetation that is suitable habitat for the intermediate snail hosts of schistosome parasites. With thousands of water reservoirs and irrigation canals developed or under construction in Africa, it is crucial to accurately assess the spatial distribution of freshwater intermediate snail hosts in rapidly changing ecosystems. Yet, standard techniques for monitoring snails are labor intensive, time consuming, and provide information limited to the small areas that are manually sampled. Consequently, in low-income countries where schistosomiasis control is needed most, large-scale programs to fight this disease generally operate with little understanding of where transmission hotspots are, and what types of environmental interventions will be most effective. As part of Stanford's Program for Disease Ecology, Health and the Environment, our research team developed a new framework [1] to assess schistosomiasis risk across large spatial scales in the Senegal River Basin (SRB) that integrates satellite data, high definition, low-cost drone imagery, and a computer vision technique, semantic segmentation. A deep learning model (U-Net) [2] was built to automatically analyze remote sensing imagery to produce segmentation maps of aquatic vegetation. Accurate and up-to-date knowledge of areas at highest risk for disease transmission can increase the effectiveness of control interventions by targeting habitat of disease-carrying snails. With the deployment of this new framework, local governments might better target e …