[ITK-users] Float dicom images

[David Clunie]

We (DICOM) have only recently added support for floating point
pixel data in DICOM (Sup 172), so I don't know if any of the
toolkits used by ITK support it yet.

I know dcmtk will, hopefully soon, but I can't speak for GDCM.

I have some samples at:

   https://www.dropbox.com/sh/n4iqdjbhmwljx5e/AABllea-Muk0C2oo576nm_Mua?dl=0

if you want to try them; not perfect yet, but they have the float pixel
data payload.

I am assuming that you actually want to write floating point pixels,
rather than integer quantized pixels with a floating point scaling
factor.

David

On 12/2/14 3:40 PM, Matias Montroull wrote:
> adding insight users..
>
> I've never dealt with float dicom images, can you send me a sample image? I
> can try to write and see what's the problem. isn's saving the image or you
> get a wrong saved image, such as WL WC changed etc?
>
> On Tue, Dec 2, 2014 at 4:53 PM, Tim Allman <dr.tim.allman at gmail.com> wrote:
>
>>
>> On Dec 2, 2014, at 14:44 , Matias Montroull <matimontg at gmail.com> wrote:
>>
>>> are you trying to write float dicom images?
>>
>> Yes. Sorry for noting that only in the subject line. :-(
>>
>>>
>>> On Tue, Dec 2, 2014 at 4:35 PM, Tim Allman <dr.tim.allman at gmail.com>
>> wrote:
>>> I have been having a bit of trouble trying to write Dicom images using
>> itk::GDCMImageIO. The code suggests that this is supported but I know that
>> this is a new feature. Has anyone else had any experience with this?
>>>
>>> Thanks,
>>> Tim