Regional spatial normalization: Toward an optimal target

Purpose: The purpose of this work was to develop methods for defining, cons tructing, and evaluating a "minimal deformation target" (MDT) brain for mul tisubject studies based on analysis of the entire group. The goal is to pro vide a procedure that will create a standard, reproducible target brain ima ge based on common features of a group of three-dimensional MR brain images . Method: The average deformation and dispersion distance, derived from discr ete three-dimensional deformation fields (DFs), are used to identify the be st individual target (BIT) brain. This brain is assumed to be the one with the minimal deformation bias within a group of MR brain images. The BIT bra in is deter-mined as the one with the minimal target quality score, our cos t function based on the deformation displacement and dispersion distance. T he BIT brain is then transformed to the MDT brain using an average DF to cr eate an optimized target brain. This analysis requires the calculation of a large number of DFs. To overcome this limitation, we developed an analysis method (the fast method) that reduces the task from order N-2 complexity t o one of order N, a tremendous advantage for large-N studies. Results: Multiscale correlation analysis in a group of 20 subjects demonstr ated the superiority of warping using the MDT target brain, made from the B IT brain, over several individual and MDT-transformed target brains also fr om the group, Conclusion: Analysis of three-dimensional DF provides a means to quickly cr eate a reproducible MDT target brain for any set of subjects. Warping to th e MDT target was shown by an independent multiscale correlation method to p roduce superior results.