[Insight-users] modify deformation field

[Ganesh Narayanasamy]

When i work with images of brain which hardly has any non-rigid component in it, i use affine transformation. However, most of the other organs need non-rigid transform and hence we use non-rigid transform for the whole image. If there are regions that undergoes only a rigid transform, the resulting deformation field should confirm that. 

Is it possible for you to crop the image into 2 regions - one that can undergo only rigid transform and another that has all the degrees of freedom to transform? That way, you would have absolute assurance that the rigid organ has a rigid deformation field.

my 2 cents...
Ganesh

************************************************************************

"What the mind of man can conceive and believe, It can achieve" - Napolean Hill

-Ganesh 
Narayanasamy,Ph.D., PostDoc-University of Kentucky Cancer Center

Earlier work: 
http://www.sciencedaily.com/releases/2007/05/070504133017.htm

************************************************************************

> Date: Thu, 6 May 2010 09:19:48 +1000
> From: Xi Liang <liangxi1986317 at hotmail.com>
> Subject: Re: [Insight-users] modify deformation field
> 
> Thank you for your suggestion, but I think I did not make myself clear in
> the last question. I was trying to modify the deformation field itself (not
> moving image) to enforce a certain set of the deformation field is equal to
> a rigid transformation. The set of deformation field need to modify is find
> from a mask in the moving image.
> 
> The motivation is, the original deformation field is a nonrigid deformation,
> however, I would like to apply it on a image that only certain part of the
> image is nonrigid. Thus I will have to modify the deformation field to allow
> only rigid deformation applied on the rigid structure of the image,
> and nonrigid deformation on the nonrigid structures. I hope this make sense.
> 
> That was the question I want to ask. However, later I realized that
> modifying the deformation field might not be the right way to enforce the
> rigid deformation. Instead, the better way is to modify the control point
> parameters (Bspline transformation) to enforce the rigid transformation,
> based on Tanner's paper in 2000. I post a question on this in another post.
> 

 		 	   		  
_________________________________________________________________
The New Busy think 9 to 5 is a cute idea. Combine multiple calendars with Hotmail. 
http://www.windowslive.com/campaign/thenewbusy?tile=multicalendar&ocid=PID28326::T:WLMTAGL:ON:WL:en-US:WM_HMP:042010_5
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20100506/f3030b05/attachment.htm>