External Email - Use Caution        

Dear FreeSurfer team!

First of all, thanks for the great tools! I have recently started to  
work with DWI data and there are a few issues I was hoping you could  
help me with.

I have recently found a post on the FSL mailing list mentioning that  
the bvecs stored in the dicoms header on Siemens scanner are corrected  
for slice angulation  
(https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=FSL;5757c231.1005). I  
assume this means that, for Siemens dicoms, the bvecs are stored in  
image orientation? Indeed, the bvecs in the dicoms are different than  
the ones in the table provided by Siemens. When looking at the  
preprocessing performed by tracula, the flag --bvec-voxel is used when  
importing the data from dicoms; isn't this be conflicting with the  
fact that the bvecs are already in image orientation?

A second issue I have is when using the bvecs extracted using tracula  
to perform tractography in trackvis/diffusion toolkit. For some stange  
reason, I have to flip the z-axis of the bvec (obtained with either  
--bvec-voxel or --bvec-scanner) to obtain sensible tractography  
results. When comparing the bvecs from tracula to the table from  
Siemens the z-axis is indeed flipped. Unfortunately, this flip issue  
also occurs in a number of other conversion software (e.g. dcm2niix,  
mrtrix). Does anyone know where this could be coming from? Note, my  
data was acquired on a Siemens SkyraFit with software version VE11C.

Finally, does anyone know what orientation (image/scanner) is expected  
for the bvecs when using dti_tracker (in diffusion toolkit)? When  
looking at the default bvec file passed through the GUI interface to  
dti_tracker, it seems that it requires bvecs in scanner orientation,  
but I would like to confirm this. Unfortunately, I did not receive any  
reply to my emails to supp...@trackvis.org. Perhaps someone on this  
list knows?

Thanks in advance!
Vincent.



_______________________________________________
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

Reply via email to