Hybrid deformable models for three-dimensional biomedical image segmentation

The authors apply hybrid deformable models to the task of automatically partitioning a medical image into visually sensible and medically plausible regions. In so doing, the authors exploit one to the fundamental strengths of deformable models; their ability to produce smooth closed object boundaries. Deformable modeling techniques can be broadly classified into two categories; boundary-based deformable models and region-based deformable models. Both of these approaches have distinct advantages and disadvantages. Here, the authors describe a hybrid deformable modeling technique which combines the advantages of both approaches and avoids many of their disadvantages. This is accomplished by first minimizing a region-based functional to obtain initial edge strength estimates. Smooth closed object boundaries are then obtained by minimizing a boundary-based functional which is attracted to the initial edge locations. The authors also discuss the theoretical advantages of this hybrid approach over existing image segmentation methods, and show how this technique can be effectively implemented and used for the segmentation of three-dimensional biomedical images. In particular, the authors demonstrate the use of this hybrid technique in identifying body outlines and lung regions in SPECT images