Hi, I cannot find any gradient table attached to DICOM files from our GE DISCOVERY MR750 (3 Tesla) and I would like to know if anyone else is struggling with this machine. I can read the Dicom header file of the DWI sequence but I cannot localize the point where this information is stored (A link to the header that I manually anonymized, removing name and year of birth is here: https://github.com/lorenzofontanelli/dm1/blob/main/header)
Can someone help me?
Thank you!
Hi @felbamato,
You can look at the MRtrix source code to check what tags we look into for gradient info for GE specifically. GE uses the tags (0019,10BB), (0019,10BC), and (0019,10BD) for the x,y,z components of the gradient vector, which I can see in your file as (0.447000, 0.275000, 0.851000). So the information is there, though bear in mind that it’s not the only place where this information might be stored, and for GE the direction vectors are stored relative to the image axes (not the scanner PCS system – unless stored using the newer standard tags).
Is MRtrix failing to read the information from these images correctly? If so, please send me a more complete dataset privately so I can investigate.
All the best,
Donald
Dear Professor Tournier,
Thank you for your kind reply. I have sent you an email with the error messages and the DICOM files.
L.