[Insight-users] Problem with writing DICOM using KeepOriginalUID

Mathieu Malaterre mathieu.malaterre at gmail.com
Thu Apr 23 16:20:30 EDT 2009


On Thu, Apr 23, 2009 at 9:57 PM, Martin Kavec <martin.kavec at gmail.com> wrote:
>
>
> On Thu, Apr 23, 2009 at 12:50 PM, Mathieu Malaterre
> <mathieu.malaterre at gmail.com> wrote:
>>
>> On Wed, Apr 22, 2009 at 10:03 PM, Martin Kavec <martin.kavec at gmail.com>
>> wrote:
>> > I have 182 files (slices) in the output directories for the both
>> > situations,
>> > KeppOriginalUID On and commeted out. Please note that the number of
>> > slices
>> > (files) omitted by Osirix varies. If I run my code multiple times slices
>> > at
>> > different positions and different number of slices is omitted. So this
>> > problem is a bit intermittent. For this particular run there were 180
>> > slices
>> > recognized.
>>
>> ok
>>
>> >> 4. The the number of SOP Instance UID in one dir is equal to the one
>> >> in the other one.
>> >
>> > Not OK. The number of unique SOPInstanceUID and
>> > MediaStorageSOPInstanceUID is
>> > correct (182) for the series with KeepOriginalOn commented out, and 181
>> > WITH
>> > KeepOriginalOn.
>> >
>> > After a while I identified the two files having the same SOPInstanceUID
>> > and
>> > MediaStorageSOPInstanceUID, I changed them to a unique ones (checked
>> > that)
>> > and now Osirix reads in 181 slices, that is one more. But still one
>> > slice is
>> > missing.
>>
>> Can you send me a couple of those SOP Instance UID. I vaguely recall
>> an issue in gdcm 1.2.4 where the generation of unique UIDs was not
>> robust.
>
> Here they go:
>
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380994675]         #
> 62,1 SOP Instance UID
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380994951]         #
> 62,1 SOP Instance UID
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380998846]         #
> 62,1 SOP Instance UID
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380998828]         #
> 62,1 SOP Instance UID
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380998817]         #
> 62,1 SOP Instance UID
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380998915]         #
> 62,1 SOP Instance UID
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380998957]         #
> 62,1 SOP Instance UID
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380999080]         #
> 62,1 SOP Instance UID
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380999003]         #
> 62,1 SOP Instance UID
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380999053]         #
> 62,1 SOP Instance UID
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380999149]         #
> 62,1 SOP Instance UID
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380999195]         #
> 62,1 SOP Instance UID
> (0008,0018) UI
> [1.2.826.0.1.3680043.2.1125.1.102857387392.2009042219380994983]         #
> 62,1 SOP Instance UID

Ok, so this is the old (buggy) implementation. On Linux system in
order to use a better uuid generator, you need to install the
developmement file for libuuid. Typically on debian you would do:

  sudo apt-get install uuid-dev

Anyway ITK CVS HEAD should have something a little more robust -hopefully-.

Thanks,
-- 
Mathieu


More information about the Insight-users mailing list