[Insight-users] gantry tilt compensation

[Erő Zsolt]

Hi, Luis

Thank you for the informations, it was really what I needed. Just
have an other question, what is the best time to do the segmentation
in the reconstruction process. Typically for non-tilted slices, I do
the segmentation before any transformation, just on the filtered
source images, but here, you said it is good to resample the result,
but for me any resampling did a terrible work when the aim was to get
a binary mask. What is the case here? Do the segmentation (1.) on the
original sheared volume and do the affine transform (2.) after, and the
resampling (3.) (is it needed for a good binary mask?) after (1-2-3), or maybe
it is better to do the affine transform on a resampled volume (1-3-2)
or just compute the matrix of the cross product of a resampling and
shearing and use this (1-23)? The purpose of the binary mask is to be
converted later to a 3D polygon object in VTK.

Thanks,
Zsolt Ero

--
Monday, August 23, 2004, 5:40:34 AM,  Luis Ibanez <luis.ibanez at kitware.com> wrote:


LI> Hi Zsolt,

LI> The gantry tilt will be equivalent to a Shearing
LI> in the Affine Transform.  A typical shearing effect
LI> in a 2D transform will generate a matrix of the form


LI>      1  A
LI>      0  1

LI> or

LI>      1   0
LI>      B   1


LI> Where "A", and "B" will be shearing factors and can
LI> be positive or negative.

LI> The AffineTransform in ITK
LI> http://www.itk.org/Insight/Doxygen/html/classitk_1_1AffineTransform.html
LI> doesn't have an explicit method for setting shearing factors.
LI> You can however setup your matrix and then pass it to the
LI> transform with the SetMatrix() method.

LI> Once you configure your Affine transform, you may want to
LI> use the ResampleImageFilter
LI> http://www.itk.org/Insight/Doxygen/html/classitk_1_1ResampleImageFilter.html
LI> in order to resample your image into a rectilinear grid.

LI> You are strongly encouraged to read the chapter on image
LI> resampling from the ITK SoftwareGuide

LI>      http://www.itk.org/ItkSoftwareGuide.pdf



LI>    Regards,


LI>      Luis



LI> -------------------
LI> Erő Zsolt wrote:
>> I am trying to write a small program that transforms a volume of
>> slices made with tilted CT head into a standard (rectangular) volume
>> array, because no algorithms support such type of volumes in ITK or
>> VTK (as long as I know). I know, there is an affine transform in ITK,
>> but don't know what to do with it. Could you help me how to use it
>> (not for coding but mathematically)? I know the tilt angle from the
>> DICOM file, but don't know how can I reverse-transform on the
>> transformed volume. Any web resource could be useful, I looked up
>> google, but found nothing usable.
>> 
>> Thanks,
>> Zsolt Ero
>> 
>> _______________________________________________
>> Insight-users mailing list
>> Insight-users at itk.org
>> http://www.itk.org/mailman/listinfo/insight-users
>>