@ARTICLE\{IMM1999-02837,
    author       = "U. Kjems and S. C. Storther and J. Anderson and I. Law and L. K. Hansen",
    title        = "Enhancing the Multivariate Signal of 15O water {PET} Studies With a New Non-Linear Neuroanatomical Registration Algorithm",
    year         = "1999",
    month        = "apr",
    keywords     = "image registration, warping, mutual information, Non-linear warping, stereo-tactic registration, inter-subject registration, voxel similarity measures.",
    pages        = "306-319",
    journal      = "{IEEE} Transactions on Medical Imaging",
    volume       = "18",
    editor       = "",
    number       = "4",
    publisher    = "",
    url          = "http://www2.compute.dtu.dk/pubdb/pubs/2837-full.html",
    abstract     = "This paper addresses the problem of neuro-anatomical registration across individuals for functional [15O]water {PET} activation studies. A new algorithm for {3D} non-linear structural registration (warping) of {MR} scans is presented. The method performs a hierarchically scaled search for a displacement field
 maximizing one of several voxel similarity measures derived from the two dimensional histogram of matched image intensities, subject to a regularizer that ensures smoothness of the displacement field. The effect of the non-line
ar structural registration is studied when it is computed on anatomical {MR} scans
and applied to co-registered [15O] water {PET} scans from the same subjects; in this experiment a study of visually guided saccadic eye movements.

The performance of the non-linear warp is evaluated using multivariate functional signal and noise measures. These measures prove to be useful for comparing different inter-subject registration approaches, e.g. affine versus non-linear.

A comparison of 12-parameter affine registration versus non-linear registration demonstrates that the proposed non-linear method increases the number of voxels retained in the cross-subject mask. We demonstrate that improved structural registration may result in an improved multivariate functional signal-tonoise ratio. Furthermore registration of {PET} scans using the 12-parameter affine transformations that align the co-registered {MR} images does not improve registration compared to 12-parameter affine alignment of the {PET} images directly."
}