[Insight-users] Image Intensities Changed During DICOM Read-Write-Read in ITK 3.20 vs. ITK 4.4.1

[Constantine Zakkaroff]

Hi Brad,

Yes, here's the minimal code:

************************************************************************
#include <cstdlib>
#include <itkImage.h>
#include <itkMetaImageIO.h>
#include <itkGDCMImageIO.h>
#include <itkGDCMSeriesFileNames.h>
#include <itkImageFileReader.h>
#include <itkImageFileWriter.h>
#include <itkStatisticsImageFilter.h>

int main(int argc, char *argv[])
{
	std::clog << "ITK Version: " << ITK_VERSION_STRING << std::endl;

	typedef itk::GDCMSeriesFileNames NamesGeneratorType;
	typedef itk::Image<short, 3> ImageType;

	NamesGeneratorType::Pointer namesGenerator = NamesGeneratorType::New();
	namesGenerator->SetUseSeriesDetails(true);
	namesGenerator->SetInputDirectory(argv[1]);

	typedef itk::StatisticsImageFilter<ImageType> StatsFilterType;
	typedef itk::ImageFileReader<ImageType> ImageReaderType;
	typedef itk::ImageFileWriter<ImageType> ImageWriterType;

	ImageReaderType::Pointer imageReader0 = ImageReaderType::New();
	imageReader0->SetImageIO(itk::GDCMImageIO::New());
	StatsFilterType::Pointer statsFilter0 = StatsFilterType::New();
	imageReader0->SetFileName(namesGenerator->GetInputFileNames()[0]);
	statsFilter0->SetInput(imageReader0->GetOutput());
	statsFilter0->Update();

	std::clog << "Stats for file: " << 
namesGenerator->GetInputFileNames()[0] << std::endl;
	std::clog <<
		"Min: " << statsFilter0->GetMinimum() << ", "
		"Max: " << statsFilter0->GetMaximum() << ", "
		"Mean: " << statsFilter0->GetMean() << ", "
		"Sigma: " << statsFilter0->GetSigma() << ", "
		"Variance: " << statsFilter0->GetVariance() << ", "
		"Sum: " << statsFilter0->GetSum()
		<< std::endl;

	ImageWriterType::Pointer imageWriter0 = ImageWriterType::New();
	imageWriter0->SetInput(imageReader0->GetOutput());
	imageWriter0->SetImageIO(itk::GDCMImageIO::New());
	std::string fileName0 = std::string(argv[2]) + "/file00.dcm";
	imageWriter0->SetFileName(fileName0);
	imageWriter0->Update();

	ImageReaderType::Pointer imageReader1 = ImageReaderType::New();
	imageReader1->SetImageIO(itk::GDCMImageIO::New());
	StatsFilterType::Pointer statsFilter1 = StatsFilterType::New();
	imageReader1->SetFileName(fileName0);
	statsFilter1->SetInput(imageReader1->GetOutput());
	statsFilter1->Update();

	std::clog << "Stats for file: " << fileName0 << std::endl;
	std::clog <<
		"Min: " << statsFilter1->GetMinimum() << ", "
		"Max: " << statsFilter1->GetMaximum() << ", "
		"Mean: " << statsFilter1->GetMean() << ", "
		"Sigma: " << statsFilter1->GetSigma() << ", "
		"Variance: " << statsFilter1->GetVariance() << ", "
		"Sum: " << statsFilter1->GetSum()
		<< std::endl;

	system("pause");

	return EXIT_SUCCESS;
}
************************************************************************

Many thanks,
Constantine

On 12/08/2013 14:48, Bradley Lowekamp wrote:
>
> Do you have the code to share so that it can be easily reproduced?
>
>
> Thanks,
> Brad
>
> On Aug 12, 2013, at 9:33 AM, Constantine Zakkaroff <mnkz at leeds.ac.uk> wrote:
>
>> Hello ALL,
>>
>> I'm wondering if anyone can explain what's going on with the changing image intensities in a read-write-read round-trip. I think it's a bug, but the issues.itk.org is unresponsive at the moment. Another bug? ;)
>>
>> Anyhow, the strange thing is that the results for the round trip read-write-read test are very different for ITK 3.20 and ITK 4.4.1.
>>
>> I use to StatisticsImageFilter to examine the intensity values. Here's what happens, when I simply read a DICOM image, print stats, save the image and read it again, followed by printing the stats:
>>
>> ITK Version: 3.20
>> Stats for original file:
>> Min: 0, Max: 2474, Mean: 169.236, Sigma: 256.719, Variance: 65904.7, Sum: 1.40371e+007
>> Stats for saved file:
>> Min: 0, Max: 2471, Mean: 167.132, Sigma: 256.125, Variance: 65600.1, Sum: 1.38626e+007
>>
>> ITK Version: 4.4
>> Stats for original file:
>> Min: 0, Max: 2474, Mean: 169.236, Sigma: 256.719, Variance: 65904.7, Sum: 1.40371e+007
>> Stats for saved file:
>> Min: 0, Max: 7446, Mean: 509.114, Sigma: 772.59, Variance: 596895, Sum: 4.22279e+007
>>
>> In both cases the code is identical. Just compiled against different versions of ITK. The images are read and written as itk::Image<unsigned short, 3>, with no processing, no changes to any aspect of the images.
>>
>> So, ITK 3.20 introduces a small change, which I'm pretty sure shouldn't be there. But in ITK 4.4.1 the scale of intensities changes three-fold.
>>
>> I'd not be surprised by a change of this magnitude if I was photocopying those images, but not when using digital software and hardware. :^) Is there something I'm not understanding here?
>>
>> None of this happens with read(DICOM)-write(MHD)-read(MHD), where the original and round-trip values are identical:
>> Stats for saved file: file0.mhd
>> Min: 0, Max: 2474, Mean: 169.236, Sigma: 256.719, Variance: 65904.7, Sum: 1.40371e+007
>>
>> Many thanks,
>> Constantine
>>
>> _____________________________________
>> 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
>