Registration of two images is difficult if large deformations are induced due to the absence of a structure in one image. We propose a penalty term that minimizes the volume of the missing structure in one image during free-form registration. The registration optimum found is based on image similarity, provided that the missing volume is minimal. We demonstrate our method on cervical MR images for brachytherapy. The intrapatient registration problem involves one image in which a therapy applicator is present and one in which it is not. Experiments show improvement of registration when including the penalty term. The improvements of surface distance and overlap of the bladder and rectum (which are close to the applicator volume) provide proof of principle of our method. © 2013 Springer-Verlag.
|Name||Lecture Notes in Computer Science|
|Workshop||5th International Workshop on Computational and Clinical Applications in Abdominal Imaging (CCAA 2013)|
|Abbreviated title||CCAA 2013|
|Period||22/09/13 → 22/09/13|
|Other||Workshop held in conjunction with the 16th International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2013)|