The principle and applicability of an Active Appearance model based on the analysis of 3D deformation fields for segmentation has been demonstrated. Accuracy and robustness remain to be thoroughly assessed but early results suggest that the AB method is as accurate as ANIMAL, while being significantly faster. Promising features of this novel approach include (1) its speed compared to locally available segmentation methods; (2) reliance on all grey-level voxels and deformation vectors as "landmarks" and hence maximum use of information; (3) fully 3D and automated; and (4) flexibility in the choice of SOI/VOI. The major constraint is the restriction to the domain of structure neighborhoods whose non-linear registration is achievable using ANIMAL for the training of the 3D Warp Model.