APPLICATION OF CURVE FITTING METHODS TO MAPPING OF TUBERCULOSIS IN EASTERN CAPE PROVINCE, SOUTH AFRICA

Abstract

There are accurate and imprecise ways for interpolating data that is regularly distributed or scattered. Nonetheless, some techniques may be used to irregular grids and others to regular grids for data interpolation. Examining the spatial distribution of illness prevalence rates and their relationships within a specific distance and direction is crucial for spatial epidemiology. This study's goal is to use 3-D curve fitting techniques to create a graphical disease map for TB prevalent patterns. In this work, the distribution patterns of tuberculosis (TB) in the Eastern Cape Province were identified, localized, and compared for smoothing using linear and biharmonic spline methods implemented in MATLAB for the geographic and graphic depiction of the disease prevalence. The datasets are typically displayed as 3D or XYZ triplets, where Z is the variable of interest—in this case, the province's TB counts—and X and Y are the spatial coordinates. For the years 2012–2015, surface and contour maps were created to show the prevalence of tuberculosis at the province level. Biharmonic interpolations demonstrated smooth surfaces with lower sum of squares errors and regular patterns in the distribution of tuberculosis cases in the province, according to the overall aspect of all the fittings.  These innovative interpolation techniques are infrequently employed in disease mapping applications, and they offer the advantage of being evaluated at subjective places as opposed to just on a rectangular grid, as is the case with the majority of conventional GIS techniques for geospatial analysis.