Bayesian model for spatial dependance and prediction of tuberculosis.

Srinivasan, R.; Venkatesan, P.

International Journal of Research in Medical and Health Sciences; 2013; 3; 1-6.

Abstract: In the Disease analysis, there is an interest to find the spatial pattern of disease in specific regions and whether this pattern has any spatial dependence. Kriging is one method used to find the spatial dependence and extrapolate the location of cases from unmeasured locations. But Bayesian Kriging would be more appropriate in the case of tuberculosis risk where we know that other factors are strong predictors in tuberculosis disease. The aim is to study the spatial pattern and spatial dependence of tuberculosis within a Chennai ward population to gain insight into the disease spread and also, to extrapolate the disease location from the unmeasured disease locations. SAS and WinBUGS software were used for spatial analysis of tuberculosis spread. Data was obtained from National Institute for Research in Tuberculosis for Chennai district. The result reveals that Kriging has significantly improved the prediction of tuberculosis risk in parts of the Chennai city. The GIS system proves to be a friendly interface for spatial and a spatial information retrieval, which supports users with all type of statistical analysis GIS model.

Keywords: Bayesian Kriging; Autocorrelation; Moran's I; Geary's C.

Back to List of publications / Home