Location: H1 Seefahrtschule, Wallstraße 40, 23560 Lübeck
Professor Keeling works at the Institute of Mathematics and Scientific Computing, Karl-Franzens-University of Graz.
Title: Image Registration for Dynamic Contrast Enhanced Magnetic Resonance Image Sequences
Abstract: Dynamic Contrast Enhanced Magnetic Resonance Imaging (DCE-MRI) is a diagnostic approach which involves to inject a bolus of contrast agent into a patient and to follow the course of the contrast agent with high temporal resolution using advanced magnetic resonance imaging techniques. To determine potential pathology the dynamic contrast agent concentration must be tracked for individual tissue cites. When imaging the abdomen for instance there is an unavoidable motion manifested in sequences because patients cannot remain still for sufficiently long. This motion must be eliminated by registering each image of the sequence to a given reference. Defining an image similarity measure for this task is challenging in the case of DCE-MRI sequences because of changing intensities and because of the new structures which emerge once higher contrast is achieved.
Also, images of the abdomen contain many gradual intensity variations, and thus they are far from being piecewise constant as implicitly assumed for many regularization models used in image processing. Since the force driving registration is stronger for intensity as opposed to edge based similarity measures, the approach considered here is to adapt intensities in local segments of the reference image to better match intensities of local segments in the template image. The segmentation is based upon a higher order model which is more suitable for piecewise smooth as opposed to piecewise constant images. The methods implemented are seen as an approximation to a higher order Mumford-Shah registration approach, about which continuing research will be reported. Finally, an approach for eliminating motion will be discussed which involves to match the entire sequence all at once to a derived sequence.
Contact: Prof. Dr. Jan Modersitzki