A Multicomponent Approach to Nonrigid Registration of Diffusion Tensor Images
This addresses registration challenges for diffusion tensor images in medical imaging, but it is incremental as it builds on existing registration techniques.
The paper tackled nonrigid registration of diffusion tensor images by proposing a multicomponent information-theoretic measure with tensor reorientation, resulting in much better registration accuracy compared to an affine method based on mutual information in the presence of geometric distortion.
We propose a nonrigid registration approach for diffusion tensor images using a multicomponent information-theoretic measure. Explicit orientation optimization is enabled by incorporating tensor reorientation, which is necessary for wrapping diffusion tensor images. Experimental results on diffusion tensor images indicate the feasibility of the proposed approach and a much better performance compared to the affine registration method based on mutual information in terms of registration accuracy in the presence of geometric distortion.