DWI support in DICOM

[Mathieu Malaterre]

Hello again,

I am currently involved in a group which is working extensively with
DWI data (diffusion-weighted image). Their acquisition data is coming
in the form of DICOM files. Unfortunately they are currently facing two
problems:
1. Each vendors stores the gradient directions differently (if at all!)
2. Even if you have the directions, you don't have the measurement
frame(*). As far as I understand the DICOM specification, there is
currently no way to store this type of information.

For issue #1 I am currently gathering information about the major
vendors so that I can add extraction code into gdcm. This would provide
a function to extract the gradient directions and make invisible to
user the different way vendors use to store this information. For more
information see the Wiki at:
http://www.na-mic.org/Wiki/index.php/NAMIC_Wiki:DTI:DICOM_for_DWI_and_DTI

For issue #2, -again correct me if I am wrong- but I could not find
anything that could represent this measurement frame. Therefore the
solution we are using is to use an intermediate file format called
NRRD, see for instance:
http://wiki.na-mic.org/Wiki/index.php/NAMIC_Wiki:DTI:Nrrd_format
Ideally this information should be accessible from a particular
DICOM tag. Can we request an extension of the DICOM standard to allow
us to store this information directly in the DICOM file, which would
greatly simplify the process (avoiding the intermediate NRRD file
format step).

Regards,
Mathieu
(*)
"measurement frame": relationship between the coordinate frame in which
the gradient coefficients are expressed, and the physical coordinate
frame in which image orientation is defined

[dcl...@dclunie.com]

Hi Mathieu

Thanks for the very interesting links.

The proposed standard way to address this problem is with
the Enhanced MR IOD, which explicitly addresses the
attributes for encoding diffusion directionality and B-Value.

There will likely be no extensions of the existing MR IOD to
address this concern, since by policy, we obviously want
folks to use the new IOD.

It would not surprise me if in the interim, vendors started
to send some of the new Sup 49 attributes in old IOD
instances, but there will almost certainly be no official move
to standardize this, and we would certainly not consider
adding a non-Sup 49 based mechanism.

If the Sup 49 mechanisms are not sufficient (e.g. for
tensors), then we need to discuss what the gaps are
and how to fill them.

We would not, by the way, ever "encapsulate" the NRRD
header, but it would nice if there was a clear mapping between
the DICOM Sup 49 attributes and the relevant NRRD
attributes, though these are probably obvious.

David

[Mathieu Malaterre]

Thanks again David for your help !

Anyway to close the subject (hopefully!). It's great that DICOM has a
standard for this information, but until private vendors actually use
it... it's not that usefull :(

Also for future references:
#1 If the DICOM-recommended tags are used as defined, the measurment
frame is the identity matric
#2 Therefore the only deficiency is that you cannot record the full
B-matrix when it is known.

Mathieu