[Elastix] Syntax for reading raw data..?

McInnis, David dmcinnis at mteq.com
Fri May 9 16:33:04 CEST 2014


Sounds good.  Thanks for the pointers!



On Fri, May 9, 2014 at 10:05 AM, S. Klein <s.klein at erasmusmc.nl> wrote:

>  Hi David,
> It would be possible with a bit of hacking. You could extend the 2D
> transform parameter file to 3D transformation, by changing the transform
> parameters (depending on which transform you use, you need to add a number
> of "dummy parameters", such that the transformation remains the same for
> all frames/bands, and you need to modify the definitions of the resample
> domain (Size, Spacing, Origin, Index).
> Best,
> Stefan
>
>
> On 09/05/2014 14:42, McInnis, David wrote:
>
> Is there a way to have transformix  apply the transform to just a certain
> range of bands?  This is where to IO hit would be the worst if I had to
> split the cubes, apply, then reassemble.
>
> BTW, I learned that the 'raw' output is actually ENVI's format (raw plus a
> header file meant for envi).  Quite nice.
> On May 9, 2014 6:41 AM, "S. Klein" <s.klein at erasmusmc.nl> wrote:
>
>> Hi David,
>> No, it's better to split as separate 2D files. As they are just 2D files,
>> the amount of IO time should not be too much, relative to other tasks such
>> as the registration itself.
>> Kind regards,
>> Stefan
>>
>> On 08/05/2014 18:32, McInnis, David wrote:
>>
>>    Sounds like a good idea, thanks Stefan.
>>
>>  Is there a way to use a specific band in a cube as the fixed image and
>> another as the moving?
>>  Or would I need to pull the bands out as separate 2D image files?
>>
>> I'm assuming this is not possible, but thought I'd ask.
>> It's trivial to pull the slices but I'd like to take the resulting
>> transform and apply it to a certain range of bands within the cube.  Not to
>> the whole cube.
>> I could split the cube, apply the transform to one piece, and
>> reassemble..  but the IO time would double.
>>
>>  ( These are spectral cubes: spatial, spatial, spectral. )
>>
>>
>>
>>
>>
>>
>>
>>
>> On Thu, May 8, 2014 at 11:12 AM, S. Klein <s.klein at erasmusmc.nl> wrote:
>>
>>> Hi David,
>>> It is not directly possible like this.
>>> The most easy solution would be to write a little .mhd header file next
>>> to your .raw file.
>>> In the elastix example, there is an example .mhd/.raw file. Just open it
>>> with a text editor, and you will probably figure out how it works.
>>> Cheers,
>>> Stefan
>>>
>>>
>>> On 08/05/2014 17:07, McInnis, David wrote:
>>>
>>>        Hi,
>>>
>>> I'm just learning about Elastix and am trying to figure out how to read
>>> in raw data cubes.
>>>  I got as far as finding the module:  ITKIORAW   at
>>> http://www.itk.org/Doxygen45/html/group__ITKIORAW.html
>>>  but I'm not a great coder and cannot find where the syntax structure
>>> is.   And how would it translate to Elastix's syntax?
>>>
>>>  In ITK's own software guide they simply say you can read raw files..
>>> but should avoid doing so with medical data.   :-/
>>>
>>>  Anyway I tried guessing, without luck:
>>>  >   elastix -f myfile.dat 800 600 712 2   ...
>>>  meaning a 712 band cube of 800x600 as 2B ints
>>>
>>>  Anyone have a clue?
>>>
>>>  Thanks,
>>>  David
>>>
>>>
>>>  btw..  I may not be much of a coder but I do appreciate how well
>>> Elastix is written and commented and documented (ummm..  excepting my
>>> question :-)   ).    This is just beautiful work.
>>>
>>>
>>>
>>>
>>>  _______________________________________________
>>> Elastix mailing listElastix at bigr.nlhttp://lists.bigr.nl/cgi-bin/mailman/listinfo/elastix
>>>
>>>
>>> --
>>> Stefan Klein+31 10 7043442http://www.bigr.nl/people/StefanKlein
>>>
>>>
>>
>>
>> --
>> David McInnis  -  MTEQ Inc
>> 505-510-1818     dmcinnis at mteq.com
>> NVESD bldg 357, rm 225
>> Ft Belvoir, VA
>>
>>
>> --
>> Stefan Klein+31 10 7043442http://www.bigr.nl/people/StefanKlein
>>
>>
> --
> Stefan Klein+31 10 7043442http://www.bigr.nl/people/StefanKlein
>
>


-- 
David McInnis  -  MTEQ Inc
505-510-1818     dmcinnis at mteq.com
NVESD bldg 357, rm 225
Ft Belvoir, VA
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bigr.nl/pipermail/elastix/attachments/20140509/4619f797/attachment-0001.html>


More information about the Elastix mailing list