[Insight-developers] Inverse transforms

[Miller, James V (Research)]

How much of the decision of the inverse mapping is driven by the
the simplicity of metrics in the inverse mapping case?

Your metric is now:

Sum{over all p in the fixed image region} ( F(p) - M(T(p)) )^2

If the transform is changed to go from moving to fixed, the operand 
of the metric is not that different

	(F(Tm2f(q)) - M(q))^2

however, you need iterate over an non-rectangular region to compute the
metric. This sounds difficult and slow.  

It sounds like this is an engineering decision to make the registration
metrics pratical.  If so, then we just need to be very clear in the
documentation as to that the output transformation does. 

We should also provide a resample filter that will resample a moving image
to the fixed image coordinate system for display of the final results and
for transfering information from the moving image to the fixed image
coordinate frame.  For simple transforms, this will call the
ResampleImageFilter passing in the inverse of the registration transform.
For more complicated transforms, it will have to an incremental search to
establish the inverse mapping.

Jim


-----Original Message-----
From: Lydia Ng [mailto:lng at insightful . com]
Sent: Monday, November 03, 2003 1:20 AM
To: Stephen R. Aylward; Insight-developers (E-mail)
Subject: RE: [Insight-developers] Inverse transforms


Hi Stephen,

I agree with you it would be nice to address the inconsistencies.

"Inverse mapping" is important in resampling because it avoids the
problem of overlap and holes. At each iteration of image-to-image
registration you are implicitly computing the resampled image (with the
current parameters) - so wouldn't the same issues of overlap occur here
as well?

Just to clarify, currently with the "inverse mapping" scenario, the
equation for the mean squares metric looks something like this:
Sum{over all p in the fixed image region} ( F(p) - M(T(p)) )^2

So for each p, we compute mapped position T(p) using the transform and
then interpolate the moving image value at T(p).

What would be the equivalent equation and computation steps with
"forward mapping"?

The second issue is what happens when the transform is not invertible?
For example, BSplineDeformableTransform is useful for representing a
warp but there is no inverse. In this case, we can't easily flip-flop
from inverse to forward mapping or vice versa.

- Lydia


> -----Original Message-----
> From: Stephen R. Aylward [mailto:aylward at unc . edu]
> Sent: Friday, October 31, 2003 11:42 AM
> To: Insight-developers (E-mail)
> Subject: [Insight-developers] Inverse transforms
> 
> 
> Hi,
> 
> 1) ITK registration actually returns a transform from the fixed image
to
> the moving image
> 2) ITK resample image filter expects a transform from the fixed image
to
> the moving image
> 3) No model-to-image registration process is easily implemented using
a
> fixed-to-moving image transform - the model drives where in the fixed
> image the fit should be calculated.   So, all spatial-object-to-image
> transforms, the landmark-landmark registration stuff we are doing, and
> the ICP method we are implementing apply the transform to the moving
> model/points to quantify their match with the fixed image/points.
> Therefore we have to pass the Inverse transform to the resample image
> filter for them.
> 
> We have the landmark initialized MI registration app that must
> 
> 1) register using landmarks
> 2) resample the image using the inverse of that transform
> 3) invert the transform to initialize the MI registration process
> 4) resmaple the image using the non-inverse of that transform
> 
> This is really inconsistent.   Can be documented by saying most
> image-image transforms are fixed to moving transforms and all spatial
> object transforms are moving to fixed.   I say most image-image
> transforms since we are writing one that internally calculates
features
> and landmarks from the moving image to drive the registration with the
> fixed image.
> 
> I consider this a huge flaw with no easy answer - I understand why the
> resampling is driven by the fixed image.   However, couldn't the
> resampling take the inverse and all of the rest be done in the correct
> way: tranaform maps moving to fixed?
> 
> Stephen
> 
> --
> ===========================================================
> Dr. Stephen R. Aylward
> Associate Professor of Radiology
> Adjunct Associate Professor of Computer Science and Surgery
> http://caddlab . rad . unc . edu
> aylward at unc . edu
> (919) 966-9695
> 
> 
> _______________________________________________
> Insight-developers mailing list
> Insight-developers at itk . org
> http://www . itk . org/mailman/listinfo/insight-developers
_______________________________________________
Insight-developers mailing list
Insight-developers at itk . org
http://www . itk . org/mailman/listinfo/insight-developers