When I attempted to convert multi-echo mprage images using dcm2niix (version v1.0.20220720), I got the following warning messageļ¼
Chris Rorden's dcm2niiX version v1.0.20220720 (JP2:OpenJPEG) (JP-LS:CharLS) MSC1900 (64-bit Windows)
Found 960 DICOM file(s)
Slices not stacked: echo varies (TE 1.64, 3.5; echo 1, 2). Use 'merge 2D slices' option to force stacking
Warning: Assuming mosaics saved in reverse order due to 'sSliceArray.ucImageNumb'
Is this something that I can ignore? Thank you for clarifying.
Your images were acquired with H>>F image numbering mode. Siemens explicitly says you should not use this mode. Please adjust the sequence on your console. Some users assume that this acquires data in a descending rather than an ascending pattern, but this is not the case. It simply changes the order that 2D slices are written to the mosaic. I assume dcm2niix is doing the right thing, but this is virtually untested as Siemens explicitly warns users not to use this mode. Do check the spatial direction, the slice timing and the diffusion vectors if you use this mode. there is very little validation of this mode by dcm2niix (and presumably Siemens as well).
https://crnl.readthedocs.io/stc/index.ht...
see PDF by Joachim Graessner linked here https://en.wikibooks.org/wiki/SPM/Slice_...
"Do not use the mode H>>F because this complicates the numbering and you will have to sort images manually in most fMRI post-processing tools. The excitation of the slices in this case also starts caudally with the highest image numbers counting backwards!"
Hi,
I have a related question: We're using a Siemens Magnetom Skyra syngo MR XA30 scanner with a multiband acquisition. Since updating dcm2niix to version v1.0.20230411 the warning "Warning: Assuming mosaics saved in reverse order due to 'sSliceArray.ucImageNumb" occured. The images should be acquired with F >> H. When checking the slice timing provided in the json file after conversion, they seem to be correct compared to times from dicom tag 0018,9074 (as described as the last option here: https://github.com/rordenlab/dcm2niix/is...). What causes this warning and is there a way to solve it?
Thank you for your help!
Lara
Originally posted by Chris Rorden:
Hello, Your images were acquired with H>>F image numbering mode. Siemens explicitly says you should not use this mode. Please adjust the sequence on your console. Some users assume that this acquires data in a descending rather than an ascending pattern, but this is not the case. It simply changes the order that 2D slices are written to the mosaic. I assume dcm2niix is doing the right thing, but this is virtually untested as Siemens explicitly warns users not to use this mode. Do check the spatial direction, the slice timing and the diffusion vectors if you use this mode. there is very little validation of this mode by dcm2niix (and presumably Siemens as well). https://crnl.readthedocs.io/stc/index.ht... see PDF by Joachim Graessner linked here https://en.wikibooks.org/wiki/SPM/Slice_... "Do not use the mode H>>F because this complicates the numbering and you will have to sort images manually in most fMRI post-processing tools. The excitation of the slices in this case also starts caudally with the highest image numbers counting backwards!"
For Siemens systems VA-VE saving as mosaics reduced the number
and size of files created. With XA Siemens upgraded to enhanced
DICOM, with all slices from a non-mosaic image saved as a single,
compact file. Siemens considers the XA mosaics secondary capture images intended for quality
assurance only and the mosaics do not contain the rich
sequence meta-data of non-mosaic images. When you upgrade a Siemens
system to XA, it is crucial to ensure that all your sequences are
set to save enhanced, non-mosaic DICOMs. Provided with impoverished
DICOMs, dcm2niix will create impoverished NIfTIs. The solution to
this problem is to save non-mosaic enhanced images for all future
acquisitions. The Siemens Research Collaboration Manager associated
with your center can provide more details.
For background, see here:
https://github.com/rordenlab/dcm2niix/issues/236
The warning is because an operator accidentally set the image numbering of the mosaic to "H>>F", presumably assuming this would change the acquisition order from ascending to descending. However, this setting only changes the image placement in the mosaic, not the acquistion. As Siemens notes: "Do not use the mode H>>F because this complicates the numbering and you will have to sort images manually in most fMRI post-processing tools."
Thank you for the very fast reply! We double checked our
sequences and have selected F >> H for the acquisition and
ensured to export the dicoms in enhanced format. So we're not sure
what might cause this warning for our data. We would also be more
than happy to provide you with a sample data if this might help
resolve the issue? Or do you maybe have further suggestions, what
to look for?
I am unable to provide additonal insight without seeing your data. Feel free to share with my institutional email: https://sc.edu/study/colleges_schools/ar...