Chapter 11 - Applications of Statistical Deformation Model

Abstract

Statistical deformation models introduced in Chapter 2 Statistical Deformation Model: Theory and Methods come in different forms and “shapes”. They often have their own advantages and disadvantages and may seem applicable to all problems at hand. It is not an easy task in adopting or choosing from different methods. There are certain traits in the application of interest that determine best engineering strategy to approach. These traits tend to be from practical considerations, include the representation of the data, availability of the segmentation, accessibility to implementation of certain algorithms and speed of the code, to name but a few. Furthermore, both underlying representation of the deformation and the statistical methods described are not without (sometimes strong) assumptions and, in general, need to be generalized carefully to another application or even to a different source of data. Therefore, validation with respect to the application plays an important role in the field of medical image computing and remains a challenging task.