The increased interest of reducing the infection rates of neglected tropical diseases like schistosomiasis in the world has raised the necessity of developing epidemiological monitoring techniques, in order to target specific areas where the risk of infection are at highest. The aim of this project was to produce infection probability maps of the urinary schistosomiasis, caused by the parasite S.haematobium, in order to identify high risk zones where targeted interventions could be undertaken in Burkina Faso. These maps were produced thanks to Bayesian analysis techniques, using geostatistical generalized linear models. The predictions were effectuated thanks to 247 community level infection prevalence data collected from the published literature, using environmental predictors as the NDVI, population density, elevation, mean temperature, mean decadal rainfall estimates and a mean dry-season period time. The predicted results showed that prevalence rates were at highest in the northern part of the country, with a tendency to decrease in a homogeneous way to the South. The absence of heterogeneous covariates, explaining more localized environmental information like distances to water bodies or mobility information, prevented the geostatistical model to explain the local variations in S.haematobium prevalence rates. These could be integrated in the model for future works to see their capability to explain heterogeneity in the prevalence rates observations.