Obtaining a deformation model from the acquired images during treat- ment without much previous intervention or knowledge of the site and type of structures involved, able to deal with different qualities and types of images was the main motivation of creating a piecewise-non- rigid registration algorithm.
The algorithm divide the images to be register in sub-volumes (fea- turelets) and rigidly register them to the corresponding search region.
The model was first tested on a simple deformable phantom with fidu- cial markers, showing a good result in terms of the final distance of the markers obtain in comparison to rigid registration and B-spline deformable registration.
Another comparison was done using three different deformable reg- istration (DR) methods: the Demons algorithm implemented in the iP lan Software (BrainLAB AG, Feldkirchen, Germany) and two custom- developed piecewise methods using either a Normalized Correlation or a Mutual Information metric (featurelet N C and featurelet M I ). These methods were tested on data acquired using a novel purpose-built phantom for deformable registration and clinical CT/CBCT data of prostate and lung cancer patients. The Dice similarity coefficient (DSC) between manually drawn contours and the contours gener- ated by a derived deformation field of the structures in question was compared to the result obtained with rigid registration (RR).
For the phantom, the piecewise methods were slightly superior, the featurelet N C for the intramodality and the featurelet M I for the in- termodality registrations. For the prostate cases in less than 50% of the images studied the DSC was improved. Deformable registration methods improved the outcome over a rigid registration for lung cases and in the phantom study, but not in a significant way for the prostate study. A significantly superior deformation method could not be identified.