[Insight-users] Improving the BSplineDeformationTransform?

[motes motes]

On Sat, Aug 15, 2009 at 8:54 PM, Luis Ibanez <luis.ibanez at kitware.com>wrote:

> Hi Motes,
>
> Your understanding on how the BSplineDeformableTransform works
> is correct.






I am still not sure I quite understand how the deformation vectors are
placed.

I read up on some of the details in the itkSoftwareGuide. On page 40 the
concept of 'spacing' is very well illustrated and explained. Now the loop (a
bit modified from the previous example):

  RegionType::SizeType gridSizeOnImage;
  gridSizeOnImage.Fill(5);
  SpacingType spacing = fixedImage->GetSpacing();

        for(unsigned int r=0; r<Dimension; r++) {
            spacing[r] *= static_cast<double>(fixedImageSize[r] - 1)  /
static_cast<double>(gridSizeOnImage[r] - 1);
        }

actually just scales the spacing between pixel centers. If the image is a 3D
volume the spacing is a 3 component vector where each element specified the
distance/spacing between pixel centers in each dimension.

But what does this have to do with the position of the deformation vectors?

Assume that I would like to change the initial position of one of the
deformation vectors, how would that be possible?





>
>
> The locations of the nodes are never modified.
>
> The registration process only computes new values for the magnitude
> and direction of the vectors associated with the nodes.
>
> It may be possible to get more adjustable fields by moving the nodes
> as well,... but then you have to deal with a new interpolation for
> computing
> the deformation vectors in all the intermediate locations, and of course,
> you should come up with the heuristic on how to move the node locations.
>




I am thinking of using this paper as starting point:

*The adaptive bases algorithm for intensity-based nonrigid image
registration*
Rohde, G.K.; Aldroubi, A.; Dawant, B.M.

http://www.contrib.andrew.cmu.edu/~gustavor/rohde_tmi_03.pdf