Medical image processing tasks are computationally very demanding. But in a clinical environment, the time sometimes really matters. Therefore, the latest high performance architectures are used to accelerate the processing of the data.

The focus of my research is to enable and optimize of the imaging toolchain on upcoming hardware platforms. The algorithms I investigate on are reconstructions for transmission (X-ray CT) as well as emission tomography (SPECT).

Currently, I am working with Intel's Larrabee architecture which is expected to be available in late 2009 / 2010. Furthermore, I collaborate with our CUDA Guy, Benni. For further information about Larrabee I shall refer you to its Wikipedia article .

During my study thesis we worked on the development of a framework to facilitate image reconstruction (or arbitrary image processing) tasks on a variety of hardware platforms. My task was the optimization of the FFT based shift invariant filter (Shepp-Logan) for the filtered backprojection on IBM's CELL processor (CBEA).