Adaptive distance learning scheme for diffusion tensor imaging using kernel target alignment

In segmentation techniques for Di usion Tensor Imaging (DTI) data, the similarity of di usion tensors must be assessed for partitioning data into regions which are homogeneous in terms of tensor characteristics. Various distance measures have been proposed in literature for analysing the similarity of di usion tensors (DTs), but selecting a measure suitable for the task at hand is dicult and often done by trial-and-error. We propose a novel approach to semiautomatically de ne the similarity measure or combination of measures that better suit the data. We use a linear combination of known distance measures, jointly capturing multiple aspects of tensor characteristics, for comparing DTs with the purpose of image segmentation. The parameters of our adaptive distance measure are tuned for each individual segmentation task on the basis of user-selected ROIs using the concept of Kernel Target Alignment. Experimental results support the validity of the proposed method.

@InProceedings { RVTH08,
  author       = "Rodrigues, P. and Vilanova, Anna and Twellmann, T. and Haar Romeny, B.M. ter",
  title        = "Adaptive distance learning scheme for diffusion tensor imaging using kernel target alignment",
  booktitle    = "MICCAI 2008 Workshop on Computational Diffusion MRI (CDMRI), September 10th, 2008, New York USA",
  pages        = "148--158",
  year         = "2008",
  editor       = "Alexander,D. and Gee,J. and Whitaker,R.",
  url          = "http://graphics.tudelft.nl/Publications-new/2008/RVTH08"
}