Hybrid multiscale landmark and deformable image registration
An image registration technique is presented for the registration ofmedical images using a hybrid combination of coarse-scale landmark andB-splines deformable registration techniques. The technique isparticularly effective for registration problems in which the imagesto be registered contain large...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AIMS Press
2007-07-01
|
| Series: | Mathematical Biosciences and Engineering |
| Subjects: | |
| Online Access: | https://www.aimspress.com/article/doi/10.3934/mbe.2007.4.711 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | An image registration technique is presented for the registration ofmedical images using a hybrid combination of coarse-scale landmark andB-splines deformable registration techniques. The technique isparticularly effective for registration problems in which the imagesto be registered contain large localized deformations. A briefoverview of landmark and deformable registration techniques ispresented. The hierarchical multiscale image decomposition of E.Tadmor, S. Nezzar, and L. Vese, A multiscale image representationusing hierarchical $(BV,L^2)$ decompositions, Multiscale Modeling andSimulations, vol. 2, no. 4, pp. 554--579, 2004, is reviewed, and animage registration algorithm is developed based on combining themultiscale decomposition with landmark and deformable techniques.Successful registration of medical images is achieved by firstobtaining a hierarchical multiscale decomposition of the images andthen using landmark-based registration to register the resultingcoarse scales. Corresponding bony structure landmarks are easilyidentified in the coarse scales, which contain only the large shapesand main features of the image. This registration is then fine tunedby using the resulting transformation as the starting point todeformably register the original images with each other using aniterated multiscale B-splines deformable registration technique. Theaccuracy and efficiency of the hybrid technique is demonstrated withseveral image registration case studies in two and three dimensions.Additionally, the hybrid technique is shown to be very robust withrespect to the location of landmarks and presence of noise. |
|---|---|
| ISSN: | 1551-0018 |