Mjolnir: Extending HAMMER Using a Diffusion Transformation Model and Histogram Equalization for Deformable Image Registration

<table>Average Dice coefficients for different anatomical regions on the left and right hemisphere (labeled L and R) for Mjolnir registration using hard segmentation exclusively (as in HAMMER) versus using fuzzy (or soft) segmentation and histogram equalized MR intensities in the attribute vector. The top set of bars shows the average over all regions, and the error bars represent one standard deviation in each direction.</table>

International Journal of Biomedical Imaging

fig8

Figure 8

Figure 8: Mjolnir: Extending HAMMER Using a Diffusion Transformation Model and Histogram Equalization for Deformable Image Registration 