[Insight-users] itk::ImageSeriesReader - Use of ReverseOrderOn to Properly Orient Constructed Volume using a CT Data-set

[M. Wirtzfeld]

Mathieu,

Thank you for your response;  pointing me back to my original post  
and your original response.

I followed your suggestion on using the itk::GDCMSeriesFileNames  
class.  However, when I executed my code, registration did not take  
place;  the itk::RegularStepGradientDescentOptimizer class threw a  
"Gradient Magnitude Tolerance" error when using the CT data-set.  I  
obtained the same results using the MR data-set.

Suggestions?


Michael.



On 12-Aug-08, at 5:48 AM, Mathieu Malaterre wrote:

> Ref:
> http://www.itk.org/pipermail/insight-users/2008-July/026667.html
>
> itk::GDCMSeriesFileNames will sort correctly in either cases.
>
> -Mathieu
>
> On Mon, Aug 11, 2008 at 11:07 PM, M. Wirtzfeld  
> <wirtzfeld at rogers.com> wrote:
>> Good afternoon,
>>
>> I have a question regarding the use of the itk::ImageSeriesReader  
>> class.
>>
>> I have two sets of images, a CT data-set and an MR data-set.  Both  
>> data-sets
>> are of a single volume and consist of a sequence of DICOM images.
>>
>> My registration application reads a single DICOM image that acts  
>> as the
>> fixed-image (in context of ITK), while the set of DICOM images is  
>> read using
>> the aforementioned class in order to construct the volume or  
>> moving-image
>> (again in the ITK context).
>>
>> When the registration application executes using the MR data-set,
>> registration occurs.  However, when the CT data-set is used, the  
>> method
>> "ReverseOrderOn" must be called to reverse filename order,  
>> otherwise the
>> volume appears to be "flipped" and there is no overlap between the  
>> fixed-
>> and moving-images.
>>
>> I use the itk::ImageFileReader class to read the fixed-image.   
>> Given a
>> single image is read using this class, it would not need a method  
>> like
>> "ReverseOrderOn".  However, this may be the reason for the resulting
>> "flipped" volume orientation.
>>
>> Is there an inherent DICOM characteristic or tag in the DICOM  
>> necessitating
>> the need to call the "ReverseOrderOn" method?
>>
>> My immediate solution to this problem is to read the DICOM tag  
>> "Modality"
>> (0008, 0060), and depending on the type of data-set being used,  
>> call the
>> "ReverseOrderOn" method when constructing the volume.  This seems  
>> like a
>> band-aid solution - therefore this posting.  If there is a more  
>> correct
>> approach, I would certainly like to use it.
>>
>>
>> Suggestions / comment are greatly welcomed!
>>
>>
>> Thanks!
>>
>> Michael.
>>
>> Robarts Research
>> London, ON
>> Canada
>> _______________________________________________
>> Insight-users mailing list
>> Insight-users at itk.org
>> http://www.itk.org/mailman/listinfo/insight-users
>>
>
>
>
> -- 
> Mathieu

M. Wirtzfeld

"There are three kinds of lies: lies, damned lies, and statistics." -  
Benjamin Disrael





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20080812/c54792a0/attachment.html>