Sub-Saharan Africa is currently undergoing an urban demographic transition, a situation which has serious environmental and health consequences. However, there is a lack of reliable data to estimate the amount, rates, and health consequences of urban expansion on the African continent. Ghana, a relatively politically stable sub-Saharan African country became an urban majority country in 2010. The overall objectives of this research are to 1) develop and test a remote sensing approach that combine very high and moderate spatial resolution imagery in conjunction with object-based image analysis for conducting land cover/land use change (LCLUC) mapping and analysis, 2) further the understanding of urbanization in sub-Saharan Africa by comparing LCLUC of Accra and Kumasi, the two major cities in Ghana, and 3) evaluate the connections between urban LCLUC and changes in under-five mortality (5q0) for Accra's urban area. I combine geographic object-based image change analysis (GEOBICA) and a backdating approach to map and identify urban LCLUC from remote sensing images having high (FRes) and moderate (Landsat) spatial resolutions. The resulting products are then used in conjunction with spatial metrics to compare urbanization in Accra and Kumasi between 2000 and 2010. Eventually, I estimate rates of 5q0 for the neighborhoods of Accra and compare them through regression analyses to household and land cover and land use variables. The overall accuracy of the LCLUC change product derived from comparing bitemporal FRes images is 80% while that of the change product derived from comparing a time-2 FRes image to a Landsat time-1 image is 81%. Moreover, the comparison of Accra and Kumasi revealed that Kumasi grew at a rate of 6% per annum compare to 4.5% for Accra during the study period. Most new developments and landscape fragmentation in both cities took place at the urban-rural interface, within 2 km of existing built land in 2000. Finally, I found the existence of spatial variability in rates of 5q0 in Accra for the year 2000. I also found evidence of associations between high rates of under-five child mortality on one side, and non-traditional toilet types and water supply sources on the other.