||Motion Corrections in 4D Lung CT Images
||Larsen, Rasmus (Image Analysis and Computer Graphics, Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
Paulsen, Rasmus Reinhold (Image Analysis and Computer Graphics, Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
||Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark
||This thesis presents methods for modelling and reconstructing 4D lung CT.
Obtaining a good model for the lung as a function of respiration is important
for treatment of lung cancer where it is essential to radiate the entire tumor,
while limiting damaging radiation effects of healthy tissue.
The algorithmic framework used to solve this modelling is based on image registration
methods. The algorithms transform the voxels from a 3D reference
volume to sample volumes which are all positioned differently on the respiration
cycle. To achieve an optimal coordinate transformation between the reference
and sample volumes, a 4D B-spline transformation model is fitted using large
scale optimization. For this pupose two different optimizers are examined.
Following optimization, the model can be used to transform the 3D reference to
any respiration phase and is able to reconstruct CT slices within the 4D volume.
The reconstruction results are verified through visual inspection and leave out
experiments, where slices are left out to evaluate the reconstructions.
||Technical University of Denmark (DTU) : Kgs. Lyngby, Denmark
Creation date: 2009-03-04
Update date: 2009-11-04