Documentation to highlight situations or studies where multi-echo or tedana are specifically useful

[handwerkerd]

Summary

Several developers discussed adding a section to the documentation that highlights situations or specific studies that demonstrate where multi-echo fMRI in general or specific methods have been shown to be useful. This could be a either a "case studies" or "killer apps" section.

Additional Detail

I've open this issue as a place for people to suggest content that could go into this section until someone actually turns this content into a cohesive section of the documentation.

Next Steps

• Suggest case studies or papers that exist
• Suggest things that where we known multi-echo fMRI can help or thing it can help, but where you don't know of any validation studies yet.
• Compile all this info into a new section of the documentation.

[handwerkerd]

Just to get some case examples started:

• Long task designs
  Separating slow BOLD from non-BOLD baseline drifts using multi-echo fMRI
  10.1016/j.neuroimage.2014.10.051
  Can potentially distinguish slow drifts from scanner noise from slows drifts that are central to neuropharmacology or learning studies.
• Reducing dropout in ventral regions
  Making cardiac gated fMRI more realistic
  Evaluation of multi-echo ICA denoising for task based fMRI studies: Block designs, rapid event-related designs, and cardiac-gated fMRI
  10.1016/j.neuroimage.2016.07.049
  Benefits of multi-echo fMRI for brain-stem imaging and with cardiac gated fMRI.
• Paradigm free mapping methods
  Quantitative Deconvolution of fMRI Data with Multi-echo Sparse Paradigm Free Mapping
  10.1007/978-3-030-00931-1_36
  Multi-echo information helps a sparse deconvolution method become more reliable
• TSNR or CNR boosts
  8-10% boost in contrast to noise across a massively repeated task just with the optimal combination of echos and more with denoising.
  https://fim.nimh.nih.gov/presentations/effects-multi-echo-based-denoising-reliability-massively-repeated-block-design-task & https://fim.nimh.nih.gov/presentations/effect-multi-echo-denoising-amount-data-required-see-wide-spread-activity
• Non-aggressive paradigms can be accelerated slightly (or perhaps not at all) to get SNR boost from echo combination
• Something with task correlated motion, in which pure motion regressors could be harmful.

[benoitberanger]

The quest for the best: The impact of different EPI sequences on the
sensitivity of random effect fMRI group analyses
https://doi.org/10.1016/j.neuroimage.2015.10.071

In the Conclusion :

We note activation studies of the orbitofrontal cortex affected by susceptibility artifacts as an important exception, since multi-echo EPI provides superior sensitivity as compared to their single echo counterparts.

[tsalo]

@handwerkerd should we include uses of multi-echo combined with other things, like physio or phase data?

[dowdlelt]

I'm not sure where to file something like this, but this paper is interesting in using ultra high res, 7T multiecho fmri as means to examine layer profiles.
https://www.sciencedirect.com/science/article/abs/pii/S1053811911001984?via%3Dihub
One interesting finding for me was that different layers have different optimal echo times, and this can be related to myeline content.