Hi Nitrc community, 

I've acquired some single voxel spectroscopy data using a Siemens Tim Trio and am interested in matching the co-ordinates to a structural image and extracting amount of tissue type (grey, white, CSF) and voxel amounts of specific tissue types. 

I'm having some problems doing this however, as I'm unable to determine the siemen's co-ordinate system (RAS I think?) and match MRS voxel (including rotation) to structural (when I try the xy coords seem correct, but the z is WAY off).

I was thinking of transforming the MRS dicom co-ords to those used by nifti and using SPM tissue segmentation but without knowing the co-ord system can't really do much. 

So in short, does anyone have experience extracting tissue amounts from MRS voxel co-ords using a Siemens system and recommend either an automagical or manual method? Unfortunately I don't have access to LC Model or else I'd use BrainCSI (and have saved a portion of my sanity...)

Thanks, 

Peter

Hi Peter,

I just had to do this for a collaborator.  At least for our Siemens data we had a separate file which specified the position, row and column vectors, and dimensions of the voxel.  I still have a copy of the spec pdf file describing the header info. I used that to map the ROI to the structural image to report the CSF, GM, and WM proportions. 

Nick Tustison

Hi Nick,

Thanks for getting back to me. Would it be at all possible to get a copy of the pdf? Also, did you have any problems with the z coords?  

Peter

Sure.  You can download it from

https://docs.google.com/file/d/0B4SvObeE...

Also, I didn't have any problem with the z-coords.

Nick

Excellent! Thanks very much for that. 

Peter

Hi Nick,
I'm an MRI engeneer in France, and I have the same problem. Because it is too late for the link, could you repost?

Kind regards,

Laurent

No problem.  Here you go.

https://drive.google.com/file/d/0B4SvObe...

Nick

Thank you very much Nick for your document. Unfortunately my Siemens scanner is a 3T verio, operating with MRB17 software version. As a result I only have a single file (.rda), different from you with two separate files. Do you know how to read the voxel coordinates and orientation for this scanner?
Kind regards

Sorry, I don't.

You may want to check out SIVIC

http://sourceforge.net/apps/trac/sivic/

If I remembered right, you can open the .rda file with any text editor and you will see the information you need.

Hi Peter,
I used that to map the ROI to the structural image to report the CSF, GM, and WM proportions.

Excellent! Thanks very much for that.

Reviewsmag

Hi again, 

I've managed to get the co-registration working fine using rda files, but am having a problem using the .ima files straight from the magnet. 

Most of the information is translatable, but in the rda the value for the tag VOIPOSTRA and the corresponding value in the .ima files are different. Does anyone know how the rda value is calculated?

Dear Colleagues,

I encountered some problems during relocation of the MRS voxel from siemens *.rda files on my structural images, I can open the header and see the information, but I just don't get it to work. I would appreciate it so much having a code which creates a mask from the *.rda file for use with spm...

Thanks for any help!

Toby

Dear Colleagues,
I have spend a few hours thinking how to convert rda (dicom world) to nifti and don't want You to waste your time too! If You would like to convert Yours' RDAs to NIFTI masks please check out my rda2 nifti script for MATLAB.
Bartosz Kossowski

thank you very much

Hi Laurent,
I'm an MRI engeneer in France, and I have the same problem that you for obtain the mrs coordinate of the mrs voxel in .rda.
have you find some solutions.
Kind regards,
Carole

Hello Bartosz,

Your rda2nifti script is fantastic thanks for sharing! Do you know how one might derive the MRS voxel mask from the dicom files instead of the .rda files? I have collected MRS data from a Siemens Trim Trio as well. Due to the multi-TE acquisition, we went through the spectro-DICOM option which outputs the whole TE-stepped series in Siemens MRS DICOM format rather than rda format. Is there any way to extract this same information that the matlab script calls for via the dicoms?

I would appreciate any guidance you can offer!

Thanks,

Emily