NITRC dcm2nii Forum: questions

RE: Option -f %H seems to trip over spaces in StudyDescription in version 1.0.20260724

Paul Wright — Thu, 24 Sep 2026 15:19:35 GMT

I've tried editing my StudyDescription to only alphanumeric, but I still get the output in subdirectory "Unknown", all files together, filenames using series number and series description.

Option -f %H seems to trip over spaces in StudyDescription in version 1.0.20260724

Paul Wright — Wed, 23 Sep 2026 16:06:48 GMT

I have a study with spaces in the StudyDescription. Using dcm2niix v1.0.20260416, with `-f %H` it organised files as expected, but in v1.0.20260724 it seems to have split study directories on space, and then reverted to default file naming for the converted files.
Example:
StudyDescription = "The Example Study^TES"
20260416 creates:
./The\ Example\ Study/TES/sub-$PatientID/ses-1/anat/sub-$PatientID_ses-1_acq-p2_run-$SeriesNumber_T1w.nii
20260724 creates:
./The/Example/Study/TES/Unknown/$SeriesNumber_$SeriesDescription
0416 organises images into BIDS datatype directories. 0724 puts all the images together in the "Unknown" directory. Looking at a sidecar from 0724, it has the same values for "BidsGuess" as 0416.
Using `-f %h` gives similar results for both versions, but 0724 uses PatientID for sub, whereas 0416 just uses "1". Sorting of scans into datatypes is the same for both.
I realise these functions are under development so I'm using advisedly, but thought this might be useful feedback.

RE: GLIBC version error

Paul Wright — Wed, 23 Sep 2026 10:03:54 GMT

Hi Chris
Thanks. This worked.
On my university's cluster I had to:
<ol>
<li>Install cmake locally (download, extract, copy to `~/.local/bin` etc.)</li>
<li>Activate modern gcc with `scl enable devtoolset-11 bash` (your sysadmin might choose their own scl collection name, not devtoolset-n)</li>
</ol>
In case that helps anyone else.

RE: Siemens XA images exported as classic dicom

Jason Craggs — Tue, 22 Sep 2026 15:59:18 GMT

I apoligize for the ambiguity. 
Therer are several scan types in the data set.
T1
fMRI x 3
Resting state x 3
DTI weighted scan.
The data I got had been preprocessed via spm. I ran the 1st level analysis on the fMRI data and no signifigant voxels survived the analysis for the vast majority of scans. So I started from scratch.
I have now converted the images 3 diferent ways with seemingly different results (samples below). I just want to make sure that the data are converted correctly and are not the source of my problems.
Any insights or suggestion are appreciated.
 
#1 /Applications/MRIcron.app/Contents/Resources/dcm2niix -f "%f_%p_%t_%s" -p y -z y "/Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1" Chris Rorden's dcm2niiX version v1.0.20190902  (JP2:OpenJPEG) (JP-LS:CharLS) Clang10.0.1 (64-bit MacOS) Found 719 DICOM file(s) slice orientation varies (localizer?) [0 1 0 0 0 -1] != [1 0 0 0 0 -1] Warning: interpolated protocol 'DTI' may be unsuitable for dwidenoise/mrdegibbs. /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/DICOM/19041901/44230000/71606048 Warning: interpolated protocol 'DTI' may be unsuitable for dwidenoise/mrdegibbs. /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/DICOM/19041901/44230000/71607126 Warning: interpolated protocol 'DTI' may be unsuitable for dwidenoise/mrdegibbs. /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/DICOM/19041901/44230000/71607137 Warning: interpolated protocol 'DTI' may be unsuitable for dwidenoise/mrdegibbs. /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/DICOM/19041901/44230000/71607148 Warning: interpolated protocol 'DTI' may be unsuitable for dwidenoise/mrdegibbs. /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/DICOM/19041901/44230000/71607159 Skipping non-image DICOM: /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/DICOM/19041901/44230000/71607170 Warning: interpolated protocol 'DTI' may be unsuitable for dwidenoise/mrdegibbs. /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/DICOM/19041901/44230000/71607170 Warning: interpolated protocol 'DTI' may be unsuitable for dwidenoise/mrdegibbs. /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/DICOM/19041901/44230000/71607181 Convert 120 DICOM as /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/SPIN2_052_V1_fMRI_(Pain_2)_20190418171025_21 (76x76x36x120) Compress: "/usr/local/bin/pigz" -b 960 -n -f -6 "/Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/SPIN2_052_V1_fMRI_(Pain_2)_20190418171025_21.nii" Warning: Slice timing appears corrupted (range 10800..14267.5, TR=3600 ms)
 #2 dcm2niix -b y -z y -o /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted -f ${sub}_%p_s%2s ./SPIN2_052_V1                                                               11:56AM Chris Rorden's dcm2niiX version v1.0.20250505  Clang15.0.0 x86-64 (64-bit MacOS) Found 719 DICOM file(s) Error: DICOM incompatible with NIfTI slice orientation varies (issue 894, localizer?) [0 1 0 0 0 -1] != [1 0 0 0 0 -1] Skipping non-image DICOM: ./SPIN2_052_V1/DICOM/19041901/44230000/71607170 Warning: 4D Siemens XA images should be exported as enhanced not classic DICOM. Slice times and other properties may be inaccurate. Convert 120 DICOM as /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_fMRI_(Pain_2)_s21 (76x76x36x120) Compress: "/usr/local/bin/pigz" -b 960 --no-time -n -f -6 "/Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_fMRI_(Pain_2)_s21.nii" Convert 1 DICOM as /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_DTI_s09 (244x244x32x1) Compress: "/usr/local/bin/pigz" -b 960 --no-time -n -f -6 "/Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_DTI_s09.nii" Convert 1 DICOM as /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_DTI_s06 (244x244x32x1) Compress: "/usr/local/bin/pigz" -b 960 --no-time -n -f -6 "/Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_DTI_s06.nii" Warning: 4D Siemens XA images should be exported as enhanced not classic DICOM. Slice times and other properties may be inaccurate. Convert 120 DICOM as /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_fMRI_(resting_1)_s13 (76x76x36x120) Compress: "/usr/local/bin/pigz" -b 960 --no-time -n -f -6 "/Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_fMRI_(resting_1)_s13.nii" Convert 1 DICOM as /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_SAG_T2_SPACE_s03 (256x256x176x1) Compress: "/usr/local/bin/pigz" -b 960 --no-time -n -f -6 "/Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_SAG_T2_SPACE_s03.nii" Convert 1 DICOM as /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_DTI_s11 (244x244x32x1) Compress: "/usr/local/bin/pigz" -b 960 --no-time -n -f -6 "/Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_DTI_s11.nii" Warning: 4D Siemens XA images should be exported as enhanced not classic DICOM. Slice times and other properties may be inaccurate. Convert 120 DICOM as /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_fMRI_(Pain_3)_s23 (76x76x36x120) Compress: "/usr/local/bin/pigz" -b 960 --no-time -n -f -6 "/Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/052_converted/_fMRI_(Pain_3)_s23.nii" Warning: 4D Siemens XA images should be exported as enhanced not classic DICOM. Slice times and other properties may be inaccurate.
 #3 dcm2niix -b y -z y -o /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/dcm2niix_v2 /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/                 11:39AM Chris Rorden's dcm2niiX version v1.0.20260724  Clang15.0.0 x86-64 (64-bit MacOS) Found 1438 DICOM file(s) 120 images have identical time, series, acquisition and instance values. DUPLICATES REMOVED. Convert 120 DICOM as /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/dcm2niix_v2/SPIN2_052_V1_fMRI_(Pain_2)_20190418171025_21 (76x76x36x120) 1 images have identical time, series, acquisition and instance values. DUPLICATES REMOVED. Convert 1 DICOM as /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/dcm2niix_v2/SPIN2_052_V1_DTI_20190418171025_9 (244x244x32x1) 1 images have identical time, series, acquisition and instance values. DUPLICATES REMOVED. Convert 1 DICOM as /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/dcm2niix_v2/SPIN2_052_V1_DTI_20190418171025_6 (244x244x32x1) 120 images have identical time, series, acquisition and instance values. DUPLICATES REMOVED. Convert 120 DICOM as /Volumes/Data3/imaging/McCrae/Hellbender/sleeplab_1/SPIN2_052_V1/dcm2niix_v2/SPIN2_052_V1_fMRI_(resting_1)_20190418171025_13 (76x76x36x120) 1 images have identical time, series, acquisition and instance values. DUPLICATES REMOVED.

RE: GLIBC version error

Chris Rorden — Tue, 22 Sep 2026 15:02:24 GMT

Paul
  Yes, dcm2niix is benefit from a more modern glib. It does have very few dependencies. Therefore, I suggest you roll your own.
 
git clone https://github.com/rordenlab/dcm2niix.git cd dcm2niix mkdir build && cd build cmake .. make

RE: Siemens XA images exported as classic dicom

Chris Rorden — Tue, 22 Sep 2026 13:48:18 GMT

Jason
Your question is underspecified, as I have no idea what type of data it is in terms of number of participants, sessions or modalities. Here are two thoughts:
1. I often use the dcm2niix "rename" function to renaming DICOMs and place them into hierarchical folders, the command I use is 
dcm2niix -r y /path/to/dicoms
which is the same as dcm2niix -r y -f %t/%s_%p/%4r_%o.dcm /path/to/dicoms So the new output name is dased on the series date/time, series number, protocol name, instance number and media UID: https://github.com/rordenlab/dcm2niix/blob/master/FILENAMING.md 2. If your DICOMs are well behaved (e.g. reproin), you could try to see if BIDSvue can currate for you https://bidsvue.org/

Siemens XA images exported as classic dicom

Jason Craggs — Mon, 21 Sep 2026 17:32:28 GMT

Greetings, 
I am in desperate need of help.
I have a large data set that was all exported as classic dicom data. The data were cleared off the pacs server. What can I do to make the data useable? I appreciated any and all suggestions.
 
Cheers, Jason

GLIBC version error

Paul Wright — Mon, 21 Sep 2026 14:32:52 GMT

<div class="x_elementToProof" data-olk-copy-source="MessageBody">Starting with dcm2niix v1.0.20260724, I am getting this error:</div>
<div class="x_elementToProof"> </div>
<div class="x_elementToProof">./dcm2niix: /lib64/libm.so.6: version `GLIBC_2.29' not found (required by ./dcm2niix)</div>
<div class="x_elementToProof"> </div>
<div class="x_elementToProof"> </div>
<div class="x_elementToProof"> </div>
<div class="x_elementToProof">I am running on a shared cluster based on CentOS 7.9, so I realise it's likely because the distro is outdated. I have asked our cluster admin for a workaround. I am starting this thread in case anyone else encounters the same issue and will post my workaround when and if I get it working.</div>
<div class="x_elementToProof"> </div>
<div class="x_elementToProof">If you already have a fix, please post it here!</div>
<div class="x_elementToProof"> </div>
<div class="x_elementToProof">v1.0.20260416 works on my cluster.</div>
<div class="x_elementToProof"> </div>
<div class="x_elementToProof">Paul Wright</div>

RE: Extracting TotalReadOutTime from Siemens XB DICOM files

Chris Rorden — Fri, 31 Jul 2026 15:52:09 GMT

1. XB10 is a new release. Can you make sure you are running v1.0.20260724 which was validated on XB10 data (albeit a 3T Vida).   https://github.com/rordenlab/dcm2niix/releases 2. It might be worth knowing the sequence where readout time is not provided. In particular, you will need the latest version of dcm2niix to read details from the ep3d research sequences.    https://github.com/neurolabusc/dcm_qa_ep3d 3. If thse do not resolve your problem, I suggest creating a Github issue and providing steps to replicate.

Extracting TotalReadOutTime from Siemens XB DICOM files

Stuart O'Connor — Fri, 31 Jul 2026 14:30:23 GMT

Dear dcm2nii Team,
We have recently upgraded our scanner to a Siemens 7T Terra X with the new XB software Platform.
Unfortuantely dcm2nii does not appear to extract information from the DICOM on TotalReadOutTime (and/or Effective Echo Spacing + ReconMatrixPE for it's calculation) for use with FSL Topup.
I was wondering if there was a solution to this or whether this was a known issue?
Any help would be greatly appreciated
Best Wishes,
Dr Stuart O'Connor