In order to validate the algorithm with real data, eleven subjects were scanned on a Philips Gyroscan ACS 1.5 Tesla superconducting magnet system at the Montreal Neurological Institute using a T1-weighted 3-D spoiled gradient-echo acquisition with sagittal volume excitation (TR=18, TE=10, flip angle=30o, 1mm isotropic voxels, 140-180 sagittal slices). Without loss of generality, one of the 11 MRI volumes was chosen to serve as the target and the other 10 were registered to it.