Biomedical Engineering Reference
In-Depth Information
medical image segmentation with partly missing boundaries and subjective con-
tour extraction were discussed. The method was presented for 2D image seg-
mentation. However, as is common in level set methods, the extension to 3D
case is straightforward and can be done easily using ideas of this chapter.
11.6 Acknowledgements
This work was supported by NATO Collaborative Linkage Grant No. PST.CLG.
979123. The work of the first author was supported by the grant VEGA 1/0313/03,
and by the European project “Visual Modeling” in Stefan Banach International
Mathematical Centre PAN and ICM, Warsaw University. This work was partially
supported by MIUR, grant number 2002013422-005.
Questions
1.
Outline the level set segmentation models used in the last decade. What is
an advection-diffusion mechanism in such models?
2.
What is the difference between previous level set segmentation models and
Riemannian mean curvature flow of graphs discussed in this chapter?
3.
What are the main principles and advantages of the semi-implicit time
discretization?
4.
How is the segmentation partial differential equation (11.2) discretized
by the co-volume method?
5.
What are the differences between semi-implicit co-volume method and ex-
plicit finite difference method?
6.
What are the properties of the system matrix given by the semi-implicit
co-volume scheme?
7.
How can you get unconditional stability of the semi-implicit co-volume
level set method?
8.
What are the efficient methods for solving linear systems arising in the
semi-implicit co-volume level set method?
