Curvature Based Registration with Applications to MR-Mammography

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 O(nlogn) 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.

[1]  C. Broit.  Optimal registration of deformed images , 1981 .

[2]  Balraj Naren,et al.  Medical Image Registration , 2022 .

[3]  Sven Kabus,et al.  B-spline registration of 3D images with Levenberg-Marquardt optimization , 2004, SPIE Medical Imaging.

[4]  Oliver Schmitt,et al.  Registrierung einer hochaufgelösten histologischen Schnittserie eines Rattenhirns , 2001, Bildverarbeitung für die Medizin.

[5]  Karl Rohr,et al.  Landmark-Based Image Analysis , 2001, Computational Imaging and Vision.

[6]  Stefan Heldmann,et al.  Fast Computation of Mutual Information in a Variational Image Registration Approach , 2004, Bildverarbeitung für die Medizin.

[7]  Yali Amit,et al.  A Nonlinear Variational Problem for Image Matching , 1994, SIAM J. Sci. Comput..

[8]  Jan Modersitzki,et al.  Fast image registration - a variational approach , 2003 .

[9]  Jan Modersitzki,et al.  Fast inversion of matrices arising in image processing , 1999, Numerical Algorithms.

[10]  Jan Modersitzki,et al.  A Super Fast Registration Algorithm , 2001, Bildverarbeitung für die Medizin.

[11]  Paul Suetens,et al.  Free-Form Registration Using Mutual Information and Curvature Regularization , 2003, WBIR.

[12]  Gerard L. G. Sleijpen,et al.  Differences in the Effects of Rounding Errors in Krylov Solvers for Symmetric Indefinite Linear Systems , 2000, SIAM J. Matrix Anal. Appl..

[13]  G. M.,et al.  Partial Differential Equations I , 2023, Applied Mathematical Sciences.

[14]  Jan Modersitzki,et al.  Combination of automatic non-rigid and landmark based registration: the best of both worlds , 2003, SPIE Medical Imaging.

[15]  Ruzena Bajcsy,et al.  Multiresolution elastic matching , 1989, Comput. Vis. Graph. Image Process..

[16]  Jan Modersitzki,et al.  FLIRT: A Flexible Image Registration Toolbox , 2003, WBIR.

[17]  Oliver Schmitt,et al.  Elastic registration of brain images on large PC-Clusters , 2001, Future Gener. Comput. Syst..

[18]  Jan Modersitzki,et al.  Fast Curvature-Based Registration of MR Mammography Images , 2002, Bildverarbeitung für die Medizin.

[19]  Morten Bro-Nielsen,et al.  Fast Fluid Registration of Medical Images , 1996, VBC.

[20]  Jan Modersitzki,et al.  Optimal Image Registration with a Guaranteed One-to-one Point Match , 2003, Bildverarbeitung für die Medizin.

[21]  Oliver Schmitt,et al.  Effiziente, nicht-lineare Registrierung eines histologischen Serienschnittes durch das menschliche Gehirn , 2001, Bildverarbeitung für die Medizin.

[22]  B. Fischer,et al.  Visual field representations and locations of visual areas V1/2/3 in human visual cortex. , 2003, Journal of vision.

[23]  Oliver Schmitt,et al.  Elastic Matching of Very Large Digital Images on High Performance Clusters , 1999, HPCN Europe.

[24]  J. Modersitzki,et al.  Combining landmark and intensity driven registrations , 2003 .