Translator Disclaimer
29 April 2005 Interactive constraints for 3D-simplex meshes
Author Affiliations +
Medical image segmentation is still a very time consuming task and therefore not often integrated into clinical routine. Various 3D segmentation approaches promise to facilitate the work. But they are rarely used in clinical setups due to complex intialization and parametrization of such models. Clinical users need interactive tools, intuitive and easy to handle. They do not want to play around with a set of parameters which will differ from dataset to dataset and often have a non-intuitive meaning. In this work new interactive constraints for deformable three-dimensional 2-simplex meshes are presented. The user can define attracting points in the original image data. These attractors are considered during model deformation and the new forces guarantee that the surface model will pass through these interactively set points. By using the constraints the model parameterization is simplified. Segmentation is started with a spherical surface model which is placed inside the structure of interest and then adapts to the boundaries. The user can directly influence the evolution of the deformable model and gets direct feedback during the segmentation process. The model deformation algorithm was implemented and integrated in ITK (Insight Segmentation and Registration Toolkit). The newly developed segmentation tool was tested on cardiac image data and MRI lung images, but is suitable for any kind of 3D and 3D+t medical image data. It has been shown that the model is less sensitive to preprocessing of the input data as well as model initialization.
© (2005) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Thomas Boettger, Tobias Kunert, Hans-Peter Meinzer, and Ivo Wolf "Interactive constraints for 3D-simplex meshes", Proc. SPIE 5747, Medical Imaging 2005: Image Processing, (29 April 2005);

Back to Top