The results of the experiment to test the effect of increasing mis-registration are shown in Fig. 6. These demonstrate that, for all sizes of shuffle neighbourhood, the specificity and generalisation values increase (get worse) with increasing mis-registration. The results for different sizes of shuffle neighbourhood demonstrate that the range of mis-registration over which distinct values of specificity and generalisation are obtained increases as the neighbourhood size increases.
The results of the comparison between three different methods of NRR are shown in Fig. 9. These show that, particularly in terms of specificity, we can distinguish between the three approaches, with the fully groupwise method performing best, as anticipated. A model built using this approach is shown in Fig. 8.
Fig. 7. The first mode of an appearance model of the brain whose training set was subjected to deformation. standard deviations are shown.
Fig. 8. Appearance model which was built automatically by group-wise registration. First mode is shown, standard deviations.
Fig. 9. Registration evaluation which compares 3 different registration algorithms. Specificity is shown on the left and generalisation ability on the right. Values are the mean over a wide range of modes in the model.