PDF version of this document

Evaluating Brain Registration
using Models of Appearance

概要:

Appearance models are an applicable approach to the analysis of anatomical variability. They are able to distinguish between groups, e.g. normal and diseased, as a model encapsulates the properties of a group from which it was derived. The construction of such models is closely-related to the task of registration and it requires one-to-one correspondence, which registration is able to obtain. We developed a framework which evaluates both appearance models and registration, based on the statistics of large sets of images. The framework is capable of distinguishing between good models of the brain and worse ones. Furthermore, it provides a method of validating the models and evaluating registration. It does so by measuring how well a model and its (potentially registered) data fit together. Two measures are defined which reflect on the quality of a model. The first of these - specificity - approximates the level to which data generated by the model fits data from which the model was constructed. The complementary measure - generalisation - is able to quantify 'distance' between data from which the model was constructed and model-generated data. Results show that as models degrade in quality, their specificity and generalisation ability rise, as expected. The algorithms are used to compare models of the brains, which were built automatically by independent registration approaches. This greatly helps in identifying better model construction algorithms, which are analogous to registration algorithms. The algorithm is purely data-driven and requires no manual annotation.