The domain of medical imaging is naturally moving towards methods that can represent, and account for, localuncertainties in the image data. Even so, fast and efficient solutions that take uncertainty into account are notreadily available even for common problems such as gradient estimation. In this work we present a CUDA imple-mentation of Normalized Convolution, an uncertainty-aware image processing technique, well established in thesignal processing domain. Our results show that up to 100X speedups are possible, which enables full resolutionCT images to be processed at interactive processing speeds, fulfilling demands of both efficiency and interactivitythat exist in the medical domain.