Just to provide a data point you might find useful, I have encountered 3D/4D 
data in one file on a more regular basis in the past year.  First, with Siemens 
mosaic data from a diffusion weighted study.  These typically need a separate 
step to converting them into a standard serial image set with conforming DICOM 
tags.  More common now is "enhanced" DICOM (but again I typically only 
encounter these in MR).  The relevant tags are (0020, 9057), and others  found 
within sequence (0020,9222) discussed in C.7.6.16.2.2.4 of PS3 of the DICOM 
standard.  For MR, you can see supplement 49: 
ftp://medical.nema.org/medical/dicom/final/sup49_ft.pdf.  Outside of MR, the 
use of these can be found on page 452 of this document: 
http://medical.nema.org/Dicom/2011/11_03pu.pdf.
 
These do have the potential to save quite a bit of space, as only the first 
frame contains the repeating tags of all other image slices in that frame.

Enjoy,
Justin Senseney
BIRSS/ISL/DCB/CIT/NIH
Building 12A/2015
http://dcb.cit.nih.gov/~senseneyj
http://mipav.cit.nih.gov



-----Original Message-----
From: Clarkson, Matt [mailto:[email protected]] 
Sent: Wednesday, July 11, 2012 3:06 AM
To: Graser, Bastian
Cc: mitk-users
Subject: Re: [mitk-users] DICOM : 3D data in one file.

Hi Bastian,

Thats pretty much what I was thinking of implementing. I will give it a go.
Thanks

M

On 10 Jul 2012, at 11:01, Graser, Bastian wrote:

> Hi Matt,
> 
> Afaik there is no statement in the dicom standard for having a 3D/4D volume 
> in a single DCM file. Still you sometimes encounter such files. So far we 
> only encountered them when exporting 3D/4D ultrasound data from Philips QLab 
> software. For these, we wrote a special procedure to read them. So, we only 
> support these yet.
> 
> If you have a similar special case of DCM data, you might need to add 
> another extra procedure to handle these. You can do it pretty much the 
> same way as it is done for Philips3D Ultrasound data.  (see 
> DicomSeriesReader::ReadPhilips3DDicom in mitkDicomSeriesReader.cpp)
> 
> 1.) In the method DicomSeriesReader::LoadDicom in mitkDicomSeriesReader.txx 
> perform a check to identify, if the present dcm is of your special kind. 
> There are propably meta tags&values for your special dcm format, which you 
> can check for.
> 2.) if so, call a read-function for your special dcm, which should be 
> implemented in DicomSeriesReader::ReadMatts3DDicom(..). Again, just 
> copy&paste&adapt the ReadPhilips3DDicom function.
> 
> Biggest task here is to figure out, which tags are used for all the geometry 
> information you need (spacing x, y, z, rotation, timesteps,..) since they are 
> probably put into user-defined tags. You can use some external program like 
> DCMTK's dcmdump or Dicom Inspector.Net to have a look at all tags. If you 
> send me a list of all the tags and their value in your file, I might have a 
> look. But please delete personal information first (name, bday) if existing. 
> Keep in mind that due to the nature of some images, geometry data might be 
> missing (e.g Philips 3D ultrasound data does not have an origin). In that 
> case, fill in standard values (Origin = 0,0,0; Spacing= 1,1,1; Rotation = 1, 
> 0, 0 ,1). Another challenge is to find out, in what order the data is stored, 
> so you can read it correctly out of the data-Tag. This is a little trial & 
> error. 
> 
> 
> Regards,
> Bastian
> 
> -----Ursprüngliche Nachricht-----
> Von: Clarkson, Matt [mailto:[email protected]]
> Gesendet: Mittwoch, 20. Juni 2012 10:15
> An: mitk-users
> Betreff: [mitk-users] DICOM : 3D data in one file.
> 
> Hi there,
> 
> with regard to my previous questions here: 
> http://sourceforge.net/mailarchive/forum.php?thread_name=ECE4A32F-30FC
> -4B09-9189-1671B767274D%40live.ucl.ac.uk&forum_name=mitk-users
> 
> issues 3 and 4 were not caused by tag 0020,0037 (orientation), or tag 
> 0020,0032 (patient position) being missing.  The data was in fact SPECT data, 
> where the complete 3D volume is contained within one .dcm file.
> So, does MITK support this?  I have only seen a special case for Philips 3D 
> ultrasound.
> Has anyone else had a similar problem and resolved it?
> Any suggestions on how it should be resolved?
> 
> Thanks as always.
> 
> Matt
> 
> 
> 
> ----------------------------------------------------------------------
> --------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and 
> threat landscape has changed and how IT managers can respond. 
> Discussions will include endpoint security, mobile security and the 
> latest in malware threats. 
> http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> mitk-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/mitk-users
> 



------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and threat 
landscape has changed and how IT managers can respond. Discussions will include 
endpoint security, mobile security and the latest in malware threats. 
http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
mitk-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mitk-users

------------------------------------------------------------------------------
Live Security Virtual Conference
Exclusive live event will cover all the ways today's security and 
threat landscape has changed and how IT managers can respond. Discussions 
will include endpoint security, mobile security and the latest in malware 
threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
_______________________________________________
mitk-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mitk-users

Reply via email to