Chris Gorgolewski

I just run into this. It would be great if reprozip would give a meaningful (or googlable) error instead of just hanging.

The spec is not clear how fieldmaps should exactly be used for processing since there are many opinions what is the best way to do this. It's up to the...

What about directly coregistering the BOLD images to an EPI MNI template?

I asked around for manage to get this ENIGMA EPI template which seems to be a good candidate for coregistration: https://neurovault.org/images/55594/ @oesteban should we add it to niworkflows?

Run into this with ds000001 sub-04.

Yup! We should handle this more gracefully though. On Tue, Jun 13, 2017 at 3:16 PM, Oscar Esteban wrote: > Does it look like skull-stripped? > > — > You...

It seems that this feature will require optional custom "censoring scheme". The following options (set on the commandline) would need to be introduced: - FD threshold - standardized DVARS threshold...

Additional option: - number of additional volumes to censor before each "event"

Here's a good case we can try to test this theory on: https://s3.amazonaws.com/openneuro.outputs/447d26e4f9fe6ce7b5ff851356280f6a/dae3c6ba-9b1f-471e-b54d-64c9cfc34f91/reports/sub-9_task-cueexposure_bold.html From https://openneuro.org/datasets/ds001247/versions/00001?app=MRIQC&version=65&job=5a99767bdf832f000104dd30

The most problematic aspect of MELODIC is large memory consumption.