FLIRT with rigidity-image registration with a local non-rigidity penalty

Jan Modersitzki*

*Corresponding author for this work
29 Citations (Scopus)

Abstract

Registration is a technique nowadays commonly used in medical imaging. A drawback of most of the current registration schemes is that all tissue is being considered as non-rigid (Staring et al., Proceedings of the SPIE 2006, vol. 6144, pp. 1-10, 2006). Therefore, rigid objects in an image, such as bony structures or surgical instruments, may be transformed non-rigidly. In this paper, we integrate the concept of local rigidity to the FLexible Image Registration Toolbox (FLIRT) (Haber and Modersitzki, in SIAM J. Sci. Comput. 27(5):1594-1607, 2006; Modersitzki, Numerical Methods for Image Registration, 2004). The idea is to add a penalty for local non-rigidity to the cost function and thus to penalize non-rigid transformations of rigid objects. As our examples show, the new approach allows the maintenance of local rigidity in the desired fashion. For example, the new scheme can keep bony structures rigid during registration. We show, how the concept of local rigidity can be integrated in the FLIRT approach and present the variational backbone, a proper discretization, and a multilevel optimization scheme. We compare the FLIRT approach to the B-spline approach. As expected from the more general setting of the FLIRT approach, our examples demonstrate that the FLIRT results are superior: much smoother, smaller deformations, visually much more pleasing.

Original languageEnglish
JournalInternational Journal of Computer Vision
Volume76
Issue number2
Pages (from-to)153-163
Number of pages11
ISSN0920-5691
DOIs
Publication statusPublished - 01.02.2008

Fingerprint

Dive into the research topics of 'FLIRT with rigidity-image registration with a local non-rigidity penalty'. Together they form a unique fingerprint.

Cite this