DT-MRI regularization using anisotropic tensor field filtering

In this paper, we present a new non-iterative anisotropic method to regularize vector and higher order tensor fields. The driving application for this work is filtering data from diffusion tensor imaging. Our scheme is based on a multidimensional anisotropic adaptive filtering. The method is based on first estimating the local structure of the field and, then, combining the outputs of a set of steerable filter basis responses to the tensor field, which are adaptively weighted depending on the local structure. Promising results are presented from experiments on both synthetic data and real diffusion tensor MRI.