Add acq-<label> to filename of html report

[jhauneuro]

Summary

I have multiple dwi images in my scan session. When I preprocess these data the sub-X.html report gets overwritten in the output sub-X folder.

Additional details

The benefit is to preserve separate html reports for each preprocessed dwi in a scan session.

Next steps

Adding acq- to the name of the html report could help resolve this issue and prevent overwriting. Adding ses- will also be helpful for datasets with multiple sessions.

[tsalo]

That doesn't really fit with the naming convention of other preps. Can you expand on your workflow? Do you run QSIPrep multiple times?

[jhauneuro]

Yes, I am running qsiprep multiple times for a subject because I have dwi scans with different spatial resolutions within a session requiring a different --output-resolution in the qsiprep command.

[jhauneuro]

I can also rename the html report afterwards, I've found that works.

[mattcieslak]

oh, that's unfortunate. Maybe the best approach would be to implement --output-resolution auto

[tsalo]

The HTML report is generated from scratch each time you run QSIPrep, so it should include sections corresponding to the different acquisitions. If it's not doing that, then I think we have a bug.

We have talked about having native-min and native-max options for output resolution in the past.

[jhauneuro]

oh, that's unfortunate. Maybe the best approach would be to implement --output-resolution auto

Yes, this would be perfect for my use case. I generally set it to the min native spatial res dimension of my data so I like the idea of native-min and native-max options!

I also have a different eddy config file for each dwi sequence to set the mporder parameter. I calculate that manually as per FSL guidelines (N/4 to N/2; N=number of excitations per volume), which requires knowing the number of slices and multiband factor.

[mattcieslak]

I also have a different eddy config file for each dwi sequence to set the mporder parameter. I calculate that manually as per FSL guidelines (N/4 to N/2; N=number of excitations per volume), which requires knowing the number of slices and multiband factor.

This sounds like something that should also be automatically filled in based in BIDS. Are you currently doing anything else manually?

[jhauneuro]

Just those two parameters. If we can provide options to automatically set them that would be amazing. Thanks.