open-discussion > Bvecs/bvals for nii files?
Showing 1-10 of 10 posts
Display:
Results per page:
Mar 17, 2013  01:03 PM | Matthew Macellaio
Bvecs/bvals for nii files?
Hey everybody, 
I'm trying to get tensors using FSL's DTIFIT tool, in the process described here. However, I'm starting with nii files from here, which don't have bvecs and bvals data associated with them. DTIFIT requires those files to generate the tracks. Any help is very much appreciated!

Matt
Mar 17, 2013  05:03 PM | Leonardo Cerliani
RE: Bvecs/bvals for nii files?
Hello Matthew,

in my experience, once you have the nii you're one step too forward to retrieve the bval/bvec files... These information are usually provided by the scanner technician, or are encapsulated in the raw images which come out from the scanner (e.g. PAR/REC files for Philips, and in general in the raw dicom files). So you must be able to aske them either to your technician, or to ask her/him the raw images. One simple way to get those information in a format which is already suitable to feed into dtifit is by converting the raw images into nii using the utilities provided with MRIcroN (dcm2nii and dcm2nii_gui).

In summary, as far as I know there is no way to retrieve the gradient orientation from the nii file, you must be able to go back to the raw images or to get the acquisition scheme from the technician. Regarding the bvals, indeed the only thing you need to know (in most cases) is the b-value(s) used for your DWI acquisition, and this is something the technician can see directly on the console screen. The most intense image(s) in your nii file will be the so-called b0 images, with values in the bvecs file to be set to 0, while for all the other images you will have to write down the b-value you get on the screen (again, for simple cases; in particular multi b-value acquisitions this might change).

Last but not least, in my opinion it's always better to go back to the raw images as they come out from the scanner, since in your center there might be different acquisition schemes, and you or the tecnhician might get confused. You would then end up with an acquisition scheme that you might desperately try to fit to the data until the hell freezes, without obtaining any meaningful result... So, if possible, try to go back to the original raw images.

All the best!

leonardo


--
http://about.me/leonardo.cerliani
http://www.nin.knaw.nl/research_groups/k...

Leonardo Cerliani, PhD
BCN Neuro-Imaging Center
University Medical Center Groningen
Antonius Deusinglaan 2 (room 131)
tel. +31.363.8805
9713 AW Groningen - The Netherlands
Social Brain Lab
Netherlands Institute for Neuroscience
Royal Netherlands Academy of Arts and Sciences
Meibergdreef 47, 1105 BA Amsterdam, NL
___
{o,o}
|)__)
--"-"--
http://openlibrary.org/people/leonardoce...
Mar 17, 2013  07:03 PM | Matthew Macellaio
RE: Bvecs/bvals for nii files?
Thanks for your response Leonardo. I was hoping to generate structural connectivity maps from the datasets available here via DTI analysis, so I have no access to the DICOM files. Is there a method that I am not seeing that would allow me to generate tracts and determine which tracts are connecting different ROIs?
Mar 18, 2013  08:03 AM | Niels Foit
RE: Bvecs/bvals for nii files?
Matthew,

Have you considered running a probabilistic or deterministic fibre tracking at all? I think

this might be more like what youre looking for. Conn is more suitable for fMRI functional connectivity,

whereas your analysis needs a ROI-based fibretracking. Great tools are the DTIstudio, FSL, SPMx and the

Freiburg DTI toolbox and MEDINRIA. All of these have robust fibretracking algorithms and are easy to set up.

cheers

niels
Mar 18, 2013  01:03 PM | Leonardo Cerliani
RE: Bvecs/bvals for nii files?
Hi Matt,
since apparently you cannot process your DWI data with the information you have at the moment, one alternative would be to use a fiber tracts atlas. Inside FSL you can find some of them. Of course, this alternative comes with all the pros and cons of using an atlas.

all the best,
Leonardo
Mar 18, 2013  01:03 PM | Leonardo Cerliani
RE: Bvecs/bvals for nii files?
Niels,
did you read Matt's message carefully? He doesn't have the tables for gradient orientation and b-values

leonardo
Mar 18, 2013  02:03 PM | Niels Foit
RE: Bvecs/bvals for nii files?
lenoardo, yeah youre quite right, ive overread the part with the b0 files missing...matthew, could you

clarify which kind of file format you are using in your processed files?

cheers niels
Mar 18, 2013  05:03 PM | Matthew Macellaio
RE: Bvecs/bvals for nii files?
Ok Leonardo I'll try that.
Mar 18, 2013  05:03 PM | Matthew Macellaio
RE: Bvecs/bvals for nii files?
Niels, I have NIFTI files, for functional and anatomical scans.
Jan 12, 2022  01:01 PM | bionerd
RE: Bvecs/bvals for nii files?
Originally posted by Leonardo Cerliani:
Hi Matt,
since apparently you cannot process your DWI data with the information you have at the moment, one alternative would be to use a fiber tracts atlas. Inside FSL you can find some of them. Of course, this alternative comes with all the pros and cons of using an atlas.

all the best,
Leonardo
How does one use this?