[Insight-users] B-spline deformations for image registration?

[Luis Ibanez]

Hi Motes,


        I could find a question in your email...


Are you simply looking for generic comments in the matter ?

If so,
here are some:


Having BSpline grids that do not require a regular distribution
of nodes, and that allow to have nodes with different weights
will be very useful in medical imaging, given that most of the
time we have interesting information concentrated in the center
of the image, and a lot of pixel that we don't care about towards
the periphery.


   Whether something has been published on the matter or not,...

          I would consider that to be irrelevant.



First of all,
given that most publications in our field will be irreproducible
anyways...


and Secondly,
In today's environment of originality-obsessed publishing, the fact
that something has not been published, may indeed be taken as an
indication that this is a useful method that *actually* works in
practice, and simply lacks the "appearance of novelty" that we
shamefully have got to confuse as a requirement of scientific work.



So, if you have an scenario where you think that Non-uniform BSplines
could be applicable, it is certainly worthwhile to pursue an
implementation.


Don't let the decadence of the publishing system obstruct your
research work.



     Regards,


        Luis



------------------
motes motes wrote:
> Most of the B-spline deformations in an image registration procedure I 
> have read about typically use uniform B-splines. I have seen that 
> development on non-uniform B-splines is in progress in ITK. But what 
> benefits are earned from non-uniform B-splines in image registration?
> 
> I have read an article on non-uniform spacing of the control points in 
> relation to image registration:
> 
> http://www.springerlink.com/content/0q3bfeunq4avrwdf/
> 
> But to define a non-uniform spacing of the control points one does not 
> necessarily need to use non-uniform knot-vectors.
> 
> The purpose of non-uniform knot-vectors as I understand is to have 
> different weights for each control point, but to my knowledege no 
> results with this approach has been published.
> 
> 
> ------------------------------------------------------------------------
> 
> _____________________________________
> Powered by www.kitware.com
> 
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
> 
> Please keep messages on-topic and check the ITK FAQ at: http://www.itk.org/Wiki/ITK_FAQ
> 
> Follow this link to subscribe/unsubscribe:
> http://www.itk.org/mailman/listinfo/insight-users