T2w and acq-hippo_T2w mixed up in report

[dkp]

What happened?

I ran MRIQC on data that has two T2w images: one is a partial volume hippocampus image; the other is a full volume T2w image.

What command did you use?

apptainer run --cleanenv --bind ${DATA}:/data:ro --bind ${APP_DERIV_DIR}:/outputs ${APP} \
/data /outputs participant --participant_label ${Subject} -m T1w T2w bold dwi -w ${WORK_DIR} --verbose-reports

What version of the software are you running?

Running MRIQC version 24.1.0.dev0+g3fe90466.d20240417

How are you running this software?

Singularity

Is your data BIDS valid?

Yes

Are you reusing any previously computed results?

No

Please copy and paste any relevant log output.

NA

Additional information / screenshots

It seems that separate svgs are being generated in the figures directory BUT the actual report for both the T2w and the acq-hippo_T2w are including the acq-hippo_T2w images and ignoring the regular T2w images.

A suitable test BIDS dataset with both types of T2w image is available here.

[oesteban]

I believe #1317 may have resolved this -- if not completely, it definitely has a better approach to identify BIDS entities in the input file so a solution to this problem could be easy.

Thanks for the test dataset -- will check ASAP.

[dkp]

I reran it with 24.0.1 on our HPC. It crashed. I'm attaching the slurm log and mriqc logs in case they are useful. crash-20240820-132230-dkp-UploadMetrics.a0-57081b1e-f896-441c-b618-94674e31bb7d.txt mriqc-20240820-131213_d4e4380b-e5ac-4a53-9439-969b85ac4609.log slurm-2088514.txt

[dkp]

I reran with 24.0.2 and it is much better: A separate and correct report is generated for the T2_hippo scan (partial coverage T2). Oddly, the main T2w report includes data for both the whole head T2 and the T2 hippo scan: They are correctly separated within the report. However, this means that the T2_hippo scan information is reproduced in two places, which is okay, but perhaps not hat you were aiming for?

[oesteban]

I'm quite overloaded so I'm not sure I'll be able to replicate this myself. Could you post screen captures of what you see?

[dkp]

@oesteban Here's a tar file with the results in it: https://www.dropbox.com/scl/fi/xkkia8d9ad2rktpbx4qhb/mriqc_dev.tar?rlkey=t3vlxew717q8kexcucjzc6z7t&st=lr4dmpcs&dl=0

Two T2w reports were generated: sub-CAM003_ses-01brain_T2w.html and sub-CAM003_ses-01brain_acq-hippo_T2w.html. That seems perfect.

The sub-CAM003_ses-01brain_acq-hippo_T2w.html is exactly what one would hope (it contains only results and images for this partial-volume T2w image).

However, the sub-CAM003_ses-01brain_T2w.html report contains data for both the T2w image AND the T2w-hippo image (i.e., twice as much data as should have been produced). Obviously, this is not a devastating issue as it just means the T2w-hippo data is replicated in 2 places.

I've attached a screenshot from sub-CAM003_ses-01brain_T2w.html to illustrate.