High DVARS despite not much motion

[hankim]

Summary of what happened:

Hello all,

I am relatively new to fMRI data analysis and am currently preprocessing my data with fMRIprep. I started with version 24.1.1, but I noticed that DVARS is high and its plot looks rather unusual (please see the screenshots below), even though participants remained very still during the scan. I also tested versions 25.0.0, 22.1.1, and 21.0.4, but all produced similar DVARS plots. I have also tried adding --ignore slicetiming and --ignore fieldmaps but they did not help.

I did some search and found this post. I tried downgrading to 20.2.5 and the DVARS plot looks more stable with version 20.2.5 (please see screenshots below).

High DVARS values seemingly unrelated to movement

Dear all, As a follow-up to my previous post, I wanted to share a solution I found in case anyone else encounters a similar problem. Original preprocessing was done using the version 22.2.3 and updating to the newer 24.0 version did not resolve the issue. However, reverting back to an earlier version we previously used in our lab (20.2.5) seems to have solved the problem effectively. 23.2.3: [23.2.3] 20.2.5: [20.2.5] In the earlier version (20.2.5), the signal appears more stable (lower…

Does anyone have insights into this issue? If possible, I would prefer to stick with a newer version of fMRIprep, as there have been significant improvements in the later versions.

Thank you!

Command used (and if a helper script was used, a link to the helper script or the command generated):

module load singularity

export TEMPLATEFLOW_HOME='/project/bak/software/TemplateFlow/'
export SINGULARITYENV_TEMPLATEFLOW_HOME='/templateflow'

subjectID='207'
bids_directory='/project/bak/projects/analysis/scan/bids'
output_directory='/project/bak/projects/analysis/scan/bids/derivatives'
work_directory='/scratch/bak/analysis/scan/logs/work'
singularity run \
    --cleanenv \
    -B "${TEMPLATEFLOW_HOME}:${SINGULARITYENV_TEMPLATEFLOW_HOME}" \
    /project/bak/my_images/fmriprep-24.1.1.simg \
        ${bids_directory} ${output_directory} participant \
        --participant-label ${subjectID} \
        -w ${work_directory} \
        --notrack \
        --fs-license-file /project/bak/software/freesurfer_license.txt \
	--fs-no-reconall \
        --skip_bids_validation \

Version:

25.0.0; 24.1.1; 22.1.1; 21.0.4; 20.2.5

Environment (Docker, Singularity / Apptainer, custom installation):

Singularity

Data formatted according to a validatable standard? Please provide the output of the validator:

PASTE VALIDATOR OUTPUT HERE

Relevant log outputs (up to 20 lines):

PASTE LOG OUTPUT HERE

Screenshots / relevant information:

Screenshot for run 1, from v24.1.1 (left) and v 20.2.5 (right)

---

[earl]

Hello @hankim,

That DVARS trace on the left looks heavily driven by your voxel-wise intensities post-nuisance regression. At least that’s my guess from the flattening tail and larger initial peaks. Remember “DVARS is defined as the root mean square of the temporal change of the fMRI voxel-wise signal at each time point.” So it is a noisy signal that is very dependent on the voxel intensities of each 3D volume.

I think you’re saying the 2 screenshots are from the exact same fMRI run from two separate runs of fMRIPrep, but there are many factors which could lead to that differing DVARS signal.

I would personally discourage you from relying on DVARS as a representative signal of motion, but instead rely on FD there (framewise displacement) if it’s motion you’re wanting a steady measurement of.