[Insight-users] question regarding registration outputs

[Matt McCormick]

Hi Tim,

If viewing the data with VTK or other library, deformation in the
negative y-direction should be taken with care because of a difference
in coordinate systems [1].

HTH,
Matt

[1] http://www.slicer.org/slicerWiki/index.php/Slicer:VTK_ITK_Coordinates

On Wed, Oct 9, 2013 at 5:59 PM, Tim Bhatnagar <tim.bhatnagar at gmail.com> wrote:
> Thanks Matt,
>
> It is strange then, that I can extract data from the displacement field that
> seems to represent the field that would transform the 'pre-deformation'
> image into the 'deformation' image. (There is a significant deformation in
> the 'deformation data' in the negative y-direction, which shows up nicely in
> the displacement field.
>
> Must I be doing something strange for this to happen?
>
>
> On Wed, Oct 9, 2013 at 10:31 AM, Matt McCormick <matt.mccormick at kitware.com>
> wrote:
>>
>> Hi Tim,
>>
>> No, the WarpImageFilter should be generating a 'pre-deformation'
>> look-alike.  For more information, see the Software Guide.
>>
>> HTH,
>> Matt
>>
>> On Wed, Oct 9, 2013 at 4:48 PM, Tim Bhatnagar <tim.bhatnagar at gmail.com>
>> wrote:
>> > Hello all,
>> >
>> > I use the ITK examples ImageRegistration5, ImageRegistration8 and
>> > DeformableRegistration2 (Demon's) for registration of normal and
>> > deformed
>> > data (same subject). Briefly, I do some pre-registration of my data
>> > using
>> > ImageReg5 and ImageReg8 (for gross alignment), after which I take the
>> > produced displacement fields, and use them as the 'initial displacement
>> > field' for DefReg2.
>> >
>> > In all cases, I am using my 'pre-deformation' data as my fixed source,
>> > and
>> > my 'deformation' data as my moving source.
>> >
>> > I have some fiducial markers to check the accuracy of the generated
>> > displacement field, and it looks pretty good. However, when trying to
>> > produce figures, my registration output looks much closer to my
>> > 'deformation' data, and sometimes the output and the actual
>> > 'deformation'
>> > data are offset by a magnitude.
>> >
>> > In my situation, should I be aiming to have my registration output
>> > appear
>> > similar to my 'pre-deformation' data? (I get confused reading the ITK
>> > explanation on that) Shouldn't I be able to produce the a 'deformation'
>> > look-a-like by using WarpImageFilter with my 'pre-deformation' data and
>> > my
>> > generated displacement field?
>> >
>> > Sorry this was so wordy. Any help would be much appreciated.
>> >
>> > --
>> > Tim Bhatnagar
>> > PhD Candidate
>> > Orthopaedic Injury Biomechanics Group
>> > Department of Mechanical Engineering
>> > University of British Columbia
>> >
>> > Rm 5000 - 818 West 10th Ave.
>> > Vancouver, BC
>> > Canada
>> > V5Z 1M9
>> >
>> > Ph: (604) 675-8845
>> > Fax: (604) 675-8820
>> > Web: oibg.mech.ubc.ca
>> >
>> > _____________________________________
>> > Powered by www.kitware.com
>> >
>> > Visit other Kitware open-source projects at
>> > http://www.kitware.com/opensource/opensource.html
>> >
>> > Kitware offers ITK Training Courses, for more information visit:
>> > http://www.kitware.com/products/protraining.php
>> >
>> > 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
>> >
>
>
>
>
> --
> Tim Bhatnagar
> PhD Candidate
> Orthopaedic Injury Biomechanics Group
> Department of Mechanical Engineering
> University of British Columbia
>
> Rm 5000 - 818 West 10th Ave.
> Vancouver, BC
> Canada
> V5Z 1M9
>
> Ph: (604) 675-8845
> Fax: (604) 675-8820
> Web: oibg.mech.ubc.ca