Schistosomiasis mansoni is not just a physical disease, but is related to social and behavioural factors as well. Snails of the Biomphalaria genus are an intermediate host for Schistosoma mansoni and infect humans through water. The objective of this study is to classify the risk of schistosomiasis in the state of Minas Gerais (MGThe disease is related to social and behavioural factors, particularly inadequate public and environmental sanitation and a low level of education about health in the populations involved (Doumenge et al. 1987).Once schistosomiasis risk is identified by both environmental and social factors, new computerised analytical tools, known as Geographic Information Systems and Remote Sensing Data Analysis, have been used to map epidemiological data or analyse satellite images .In Brazil, geo-processing tools have been used in the study of schistosomiasis in the states of Bahia (BA) and Minas Gerais (MG). These studies have provided risk maps for schistosomiasis infection on a municipality basis by using multiple regression analyses that include environmental features, a priori disease prevalence data and other spatial data (Bavia et al. 1999, 2008, Martins-Bedê et al. 2009).In the present paper, a standard data-mining technique, the decision tree, was used to identify the severity of disease prevalence. This technique is based on a recursive Predictive models are used to classify different samples whose values or labels are not known. In this paper, a decision tree model is used to classify the schistosomiasis prevalence risk for the whole state. Remote sensing data and spatial sociological indicators are used to map potential risk areas which are not covered by the schistosomiasis control program. This map of potential risk areas can be used by a decision maker to evaluate municipalities outside of the program that have similar environmental and social conditions to the municipalities covered by the program. Moreover, the map can serve as a guide for future disease-control efforts in the state.
MATERIALS AND METHODSStudy area -The study area was MG, Brazil. MG is 590,000 square kilometres in size and is politically divided into 853 cities; the area has a tropical climate and includes approximately 18 million inhabitants (IBGE 2008).