A method and system for selective resolution improvement in computed tomography (CT) scanning. The method includes receiving scan data representative of a scanned object from a CT scanner and reconstructing the scan data using a first algorithm to create a first set of reconstructed data. A region of interest is identified within the first set of reconstructed data. A portion of the scan data corresponding to the region of interest is reconstructed using a second algorithm to create a second set of reconstructed data. The first set of reconstructed data and the second set of reconstructed data are combined to create combined reconstructed data.