Spatial based polynomial 2D-warping-a possibility to reduce interindividual variations in functional neuroimaging?

As an approach to reduce the variation in functional brain images between different individuals we tested different methods for an elastic image deformation by polynomial warping. The polynomials vary in degree and number and distribution of the grid-points. The computations show that image warping with second degree polynomials and equally distributed grid-points increases most efficiently the similarity of shape and gray-values in the images whereas the tested higher degree polynomials are not useful.