Conflicting study identifiers

Dear all,

I’m using heudiconv (version 0.6.0) for the first time because I want to move to BIDS. Not surprisingly, I´m finding some bumps on the road for which I´d appreciate some help.

I have a multi-subject multi-session study and I can generate the for most of the subjects and sessions. However, for some, I got an error related to conflicting study identifiers. Here are the command lines and output from the terminal:

INFO: Need to process 1 study sessions
INFO: PROCESSING STARTS: {‘subject’: ‘02’, ‘outdir’: ‘/base/Nifti/’, ‘session’: ‘001’}
INFO: Processing 960 dicoms
INFO: Analyzing 960 dicoms
/src/heudiconv/heudiconv/ UserWarning: The DICOM readers are highly experimental, unstable, and only work for Siemens time-series at the moment
Please use with caution. We would be grateful for your help in improving them
import nibabel.nicom.dicomwrappers as dw
Traceback (most recent call last):
File “/opt/miniconda-latest/bin/heudiconv”, line 11, in
load_entry_point(‘heudiconv’, ‘console_scripts’, ‘heudiconv’)()
File “/src/heudiconv/heudiconv/cli/”, line 127, in main
File “/src/heudiconv/heudiconv/cli/”, line 323, in process_args
File “/src/heudiconv/heudiconv/”, line 160, in prep_conversion
File “/src/heudiconv/heudiconv/”, line 96, in group_dicoms_into_seqinfos
studyUID, file_studyUID
AssertionError: Conflicting study identifiers found [,].

Any idea(s) on how could I resolve the issue on the study identifiers?

Thank you in advance for your time.


Sometimes this error stems from a new study identifier being generated when a subject got out of the scanner mid-session and had to be re-registered. Not sure if you have any info on that? I have tried running the script here to fix those that didn’t have other problems (that this AssertionError was just indicative of).

I have one scan (for each subject) in which subjects were taken out, and put back in.
i added

… -c dcm2niix -g accession_number -b --overwrite --minmeta

But there is still the error:

AssertionError: Conflicting study identifiers found

I like to note that in all my DICOMS AccessionNumber is empty ’ ’
I did follow the suggestion to change StudyInstanceUID using pydicom and it works!
Can I ignore StudyInstanceUID using

Did you try re-running heudiconv on the changed dicoms after changing the StudyInstanceUID using pydicom? What error are you still getting?

Just want to be clear, since I’m not totally sure which part “worked”, above!

Yes, I just wanted to make sure this is the only way (changing StudyInstanceUID before running heudiconv)

I don’t know of another way, but there is apparently a fix in the works for this, see here.

@mgxd might be able to say more about other workarounds.