Non-rigid  registration is widely used in a variety of image analysis  ap-
plications. We compare two methods for assessing the performance of group-
wise  non-rigid registration algorithms. One approach, which has been  de-
scribed previously, utilizes a measure of overlap between data labels. Our
new approach exploits the fact that, given a set of non-rigidly registered
images,  a generative statistical appearance model can be constructed.  We
observe that the quality of the model depends on the quality of the regis-
tration,  and can be evaluated by comparing synthetic images sampled  from
the  model with the original image set. We derive indices of model  speci-
ficity  and  generalisation that can be used to assess  model/registration
quality. We show that both approaches detect the loss of registration as a
set  of correctly registered MR images of the brain is progressively  per-
turbed.  We  compare the sensitivities of the different methods  and  show
that,  as  well as requiring no ground truth, our new specificity  measure
provides the most sensitive approach to detecting misregistration.