Mainz 2014 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
ST: Fachverband Strahlen- und Medizinphysik
ST 2: Biomedical Imaging I
ST 2.4: Talk
Tuesday, March 25, 2014, 14:45–15:00, RW 2
Improving Image Quality in Phase-Contrast Computed Tomography through Statistical Iterative Reconstruction — •Mathias Marschner1, Dieter Hahn1, Marian Willner1, Lorenz Birnbacher1, Andreas Fehringer1, Peter Noël2, Michael Chabior1, and Franz Pfeiffer1 — 1Department of Physics (E17) & IMETUM, Chair of Biomedical Physics, Technische Universität München, 85748 Garching, Germany — 2Department of Radiology, Klinikum rechts der Isar, Technische Universität München, 81675 München, Germany
Phase-contrast X-ray computed tomography (PCCT) provides additional soft-tissue contrast in biomedical samples. However, non-ideal measurement conditions and parameters, e.g. undersampling and a limited dynamic range can introduce artefacts into the tomographic reconstruction. Furthermore, low frequency noise dominates in the images due to the differential signal. These problems cannot be solved by conventional reconstruction algorithms e.g. filtered backprojection. Recently, more advanced methods have been proposed to improve the image quality of these tomographic images. We present results from an iterative reconstruction algorithm for PCCT. This algorithm uses statistical information of the initial measurement as well as complementary information from the attenuation and dark-field signals. These are concurrently obtained with grating-based phase-contrast imaging. This additional information mitigates the effects of non-ideal measurement conditions. We demonstrate that this algorithm improves the image quality by lowering the noise level and reducing artefacts.