Curvature Based Registration with Applications to MR-Mammography

Bernd Fischer, Jan Modersitzki


We introduce a new non-linear registration model based on a curvature type regularizer. We show that affine linear transformations belong to the kernel of this regularizer. Consequently, an additional global registration is superfluous. Furthermore, we present an implementation of the new scheme based on the numerical solution of the underlying Euler-Lagrange equations. The real discrete cosine transform is the backbone of our implementation and leads to a stable and fast backslashmathcalO(n backslashlog n)algorithm, where n denotes the number of voxels. We demonstrate the advantages of the new technique for synthetic data sets. Moreover, first convincing results for the registration of MR-mammography images are presented.
Original languageEnglish
Title of host publicationComputational Science --- ICCS 2002
EditorsPeter M. A. Sloot, Alfons G. Hoekstra, C. J. Kenneth Tan, Jack J. Dongarra
Number of pages5
Place of PublicationBerlin, Heidelberg
PublisherSpringer Berlin Heidelberg
Publication date10.04.2002
ISBN (Print)978-3-540-43594-5
ISBN (Electronic)978-3-540-47789-1
Publication statusPublished - 10.04.2002
EventInternational Conference on Computational Science 2002 - Amsterdam, Netherlands
Duration: 21.04.200224.04.2002
Conference number: 96801


Dive into the research topics of 'Curvature Based Registration with Applications to MR-Mammography'. Together they form a unique fingerprint.

Cite this