Hi all,
I was wondering if it is absolutely necessary to do distortion correction for 7T fMRI images?
Due to an error in scanning, there are no fieldmap or reverse phase-encoded images available for correction. I am aware that registration-based correction is possible, but is it a must? I am doing MNI space ROI-based functional connectivity analysis.
Specifically, will it be attacked by reviewers if distortion correction is absent for a 7T fMRI study?
Thank you very much for your inputs!
Hi @Meteorxy,
Personally, I doubt a paper would be rejected for lack of SDC, all else being okay, especially if you are doing an ROI based analysis, using more forgiving parcels containing lots of voxels. If you were trying to say that a small cluster of voxels was localized to a certain anatomical region, then I would be more skeptical.
The lack of SDC means that your BOLD and T1w image will not be perfectly aligned (with largest mismatches most likely towards the orbitofrontal and temporal regions). Without knowing more about your goals, this could be relevant for your hypothesis, but that is for you to decide.
Hope this helps!
Best,
Steven
Hi,
This paper can give you some information on the usefulness of SDC for such analyses:
Togo, H.; Rokicki, J.; Yoshinaga, K.; Hisatsune, T.; Matsuda, H.; Haga, N.; Hanakawa, T. Effects of Field-Map Distortion Correction on Resting State Functional Connectivity MRI. Front. Neurosci. 2017, 11, 656. https://doi.org/10.3389/fnins.2017.00656.
The paper below compares different SDC method at 7T with no SDC. You can see the impact that SDC makes:
Schallmo, M.; Weldon, K. B.; Burton, P. C.; Sponheim, S. R.; Olman, C. A. Assessing Methods for Geometric Distortion Compensation in 7 T Gradient Echo Functional MRI Data. Hum Brain Mapp 2021, 42 (13), 4205â4223. https://doi.org/10.1002/hbm.25540.
The method below can be used in case where no fieldmap was acquired:
Yu, T.; Cai, L. Y.; Morgan, V. L.; Goodale, S. E.; Englot, D. J.; Chang, C. E.; Landman, B. A.; Schilling, K. G. SynBOLD-DisCo: Synthetic BOLD Images for Distortion Correction of FMRI without Additional Calibration Scans; preprint; Bioengineering, 2022. SynBOLD-DisCo: Synthetic BOLD images for distortion correction of fMRI without additional calibration scans | bioRxiv.
I hope this helps.
I personally think no SDC at 7T is grounds for rejection depending, like Steven wrote, on what one is doing. Are the ROIs small, e.g. subcortical sub-regions? Were they defined on an individualâs anatomy or template anatomy? Are you smoothing a lot or a little and is this a group study? Were your parameters set such that distortion was mitigated (in general 7T is worse than 3T for distortion but not always). Many factors should come into play in a reviewerâs decision. To lesson the possibility of rejection Iâd first prove it to yourself that you believe the final alignments were satisfactory (focus on overlays zoomed into your ROIs) and then present supplementary material to the reviewers that prove it to them as well. An animated gif that flips between underlay and overlay is useful, as is presenting both âsimultaneouslyâ via edge maps such as that described in Rorden et al 2024 âImproving 3D edge detection for visual inspection of MRI coregistration and alignmentâ. I realize this is an old thread but Iâm adding this for others.
MANAGED BY INCF