[Insight-developers] itkGDCMImageIO and itkImageSeriesReader problem

[Mathieu Malaterre]

Patrick,

  Very good catch ! Indeed Tag::ReadFromPipeSeparatedString &
Tag::PrintAsPipeSeparatedString do not work together. I'll fix that
ASAP.

Thanks

On Tue, Jan 18, 2011 at 5:53 PM, Patrick Cheng
<cheng at isis.georgetown.edu> wrote:
> Hi Luis,
>
> I found the problem.
>
> It seems that, the dicom tags stored in the meta dictionary is now:
>
> "(0010|0010)" instead of "0010|0010"
>
> An extra pair of brackets have been added somewhere.
>
> Should be an easy fix. Let me know what is your solution, with or without
> brackets.
>
> Patrick
>
> On 1/18/2011 11:42 AM, Patrick Cheng wrote:
>>
>> Hi Luis,
>>
>> Thank you for the answer.
>>
>> I ran the test and it could not find the dicom tags.
>>
>> Here is the link to the testing file:
>> http://dl.dropbox.com/u/77341/I0092932.dcm
>>
>>
>> -------------------------------------------------------------------------------------------------------------------------------------
>>
>> C:\Research\Bin\ITK-VS10\bin\Release>DicomPrintPatientInformation.exe
>> I0092932.dcm
>> Patient Name : NOT FOUND
>> Patient ID : NOT FOUND
>> Patient Sex : NOT FOUND
>> Patient Age : NOT FOUND
>> Study Date : NOT FOUND
>> Modality : NOT FOUND
>> Manufacturer : NOT FOUND
>> Institution : NOT FOUND
>> Model : NOT FOUND
>>
>> -------------------------------------------------------------------------------------------------------------------------------------
>>
>>
>> Patrick
>>
>> On 1/18/2011 10:44 AM, Luis Ibanez wrote:
>>>
>>> Hi Patrick,
>>>
>>>
>>> 1) Please try the example:
>>>
>>> ITK/Code/IO/DicomPrintPatientInformation.cxx
>>>
>>> (from today's Git version),
>>>
>>> To see if you can get that tag from that specific DICOM file.
>>>
>>> No changes have been made to the metadata dictionary
>>> itself, so, this may be a secondary effect from migrating
>>> to GDCM 2.0.... we need your help to track the real source
>>> of the problem.
>>>
>>>
>>>
>>> 2) Problem 2, has always been there,
>>>
>>> If you reuse a reader, you must call
>>>
>>> reader->UpdateLargestPossibleRegion()
>>>
>>> instead of simply calling
>>>
>>> reader->Update().
>>>
>>>
>>> This is because a call to Update() doesn't recompute
>>> the largest size of the image.
>>>
>>> This applies to both the ImageFileReader and the
>>> ImageSeriesReader.
>>>
>>>
>>> Luis
>>>
>>>
>>>
>>> ----------------------------------------------------------------------------------
>>>
>>> On Fri, Jan 14, 2011 at 1:49 PM, Patrick Cheng
>>> <cheng at isis.georgetown.edu> wrote:
>>>>
>>>> Hi All,
>>>>
>>>> Recent changes in ITK (I am using nightly build as of Jan 12, 2011)
>>>> broke
>>>> our code in IGSTK.
>>>>
>>>> There are two issues emerged:
>>>>
>>>>
>>>> ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>>>>
>>>> Problem 1: Fail to expose DICOM tags.
>>>>
>>>> itk::MetaDataDictionary& dict = m_ImageIO->GetMetaDataDictionary();
>>>>
>>>> tagkey = "0018|1120";
>>>>
>>>> if( itk::ExposeMetaData<std::string>(dict,tagkey, m_GantryTilt ) )
>>>> {
>>>> ..... // Will not reach here
>>>> }
>>>>
>>>> Are these code still valid for checking DICOM tags?
>>>>
>>>>
>>>> ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>>>>
>>>> Problem 2: Memory effect of the GDCMImageIO class
>>>> This happens when we ask the m_ImageSeriesReader to read a new set of
>>>> image
>>>> that is smaller than the old one, it will throw an exception:
>>>>
>>>> m_ImageIO = itk::GDCMImageIO::New();
>>>> m_ImageIO->SetGlobalWarningDisplay(this->GetGlobalWarningDisplay());
>>>> m_ImageSeriesReader = ImageSeriesReaderType::New();
>>>> m_ImageSeriesReader->SetImageIO( m_ImageIO );
>>>> ........
>>>> m_ImageSeriesReader->SetFileNames( m_FileNames->GetFileNames(
>>>> seriesUID.front().c_str()
>>>> ) );
>>>> ........
>>>> try
>>>> {
>>>> m_ImageSeriesReader->Update();
>>>> }
>>>> catch( itk::ExceptionObject& excp )
>>>> {
>>>> ....//Exception: Requested region is (at least partially) outside
>>>> region.
>>>> }
>>>>
>>>> It looks like the ImageSeriesReader needs to be reset every time it
>>>> accepts
>>>> new file names.
>>>>
>>>> ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>>>>
>>>>
>>>> Thank you,
>>>>
>>>> Patrick
>>>>
>>>>
>>>> _______________________________________________
>>>> 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://kitware.com/products/protraining.html
>>>>
>>>> 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-developers
>>>>
>>>>
>>>
>>>
>> _______________________________________________
>> 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://kitware.com/products/protraining.html
>>
>> 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-developers
>>
>>
> _______________________________________________
> 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://kitware.com/products/protraining.html
>
> 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-developers
>



-- 
Mathieu