New preprocessing paper

[arokem]

Of note: https://www.biorxiv.org/content/biorxiv/early/2020/03/26/2020.03.25.008979.full.pdf

If I understand correctly, they advocate for motion outlier removal and within-volume motion correction.

Not obvious how this generalizes to other acquisitions and populations, though

[oesteban]

Yup, I had that one pinned to check out.

[josephmje]

Based on the gifs eddy provides for outlier replacement and within-volume motion correction, it looks like this would benefit any dataset. Although, you might have more extreme artefacts in high-motion populations.

But doing within-volume motion correction might be difficult to do for some datasets because not all of the information required (or at least what eddy depends on) is stored in the json sidecar. Some scanners leave out SliceTiming.