fMRIprep fails during recon-all: could not find enough control points

Summary of what happened:

Dear Experts,

I am running one participant through fMRIprep, but it is not working. I have tried
several times, and the problem is always the same during recon-all. I am out of ideas
about what might be happening. I am using 7T fMRI. Please see the attached crash file and recon-all.log. Do you have any idea what is happening?

Many thanks,
Daniela Valério

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



export FMRIPREP=/neurospin/unicog/resources/softwares/FMRIPREP/fmriprep-22.0.2.sing
export PROJECT=/home_local/Daniela_fmri_analyses/Participants
export FREESURFER=$FREESURFER_HOME
export SOURCE=BIDS
export DEST=fmriprep
# Print the paths
echo "FMRIPREP: $FMRIPREP"
echo "PROJECT: $PROJECT"
echo "FREESURFER: $FREESURFER"
echo "SOURCE: $SOURCE"
echo "DEST: $DEST"
singularity run --cleanenv \
-B "${PROJECT}: /data" \
-B "${FREESURFER}:/freesurfer" \ ${FMRIPREP} \
"/data/${SOURCE}" \
"/data/$DEST" \
participant \
--participant-label 06\
--output-spaces T1w MNI152NLin2009cAsym\
--bold2t1w-init header\
--bold2t1w-dof 9 \
--fs-license-file "/freesurfer/license.txt" \
-w "/data/${DEST}/work_dir" \
--skip-bids-validation

Version:

22.0.2

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):

INFO: extension is mgz
	#--------------------------------------------
	#@# Intensity Normalization Thu Jul 18 11:58:17 CEST 2024
	/data/fmriprep/sourcedata/freesurfer/sub-06/mri

	 mri_normalize -g 1 -seed 1234 -mprage -noconform nu.mgz T1.mgz 

	using max gradient = 1.000
	setting seed for random number genererator to 1234
	assuming input volume is MGH (Van der Kouwe) MP-RAGE
	not interpolating and embedding volume to be 256^3...
	reading mri_src from nu.mgz...
	normalizing image...
	NOT doing gentle normalization with control points/label
	talairach transform
	 1.27830  -0.07062  -0.02792  -4.15817;
	 0.08257   1.06935   0.28636  -34.69358;
	 0.03503  -0.22377   1.19210   19.37392;
	 0.00000   0.00000   0.00000   1.00000;
	processing without aseg, no1d=0
	MRInormInit(): 
	INFO: Modifying talairach volume c_(r,a,s) based on average_305
	MRInormalize(): 
	MRIsplineNormalize(): npeaks = 3
	Starting OpenSpline(): npoints = 3
	building Voronoi diagram...
	performing soap bubble smoothing, sigma = 8...

	Iterating 2 times
	---------------------------------
	3d normalization pass 1 of 2
	error: No such file or directory
	error: MRInormFindControlPoints: could not find enough control points
	error: No such file or directory
	error: MRInormFindControlPoints failed
	Linux nautilus 5.15.0-107-generic #117-Ubuntu SMP Fri Apr 26 12:26:49 UTC 2024 x86_64 x86_64 x86_64 GNU/Linux

	recon-all -s sub-06 exited with ERRORS at Thu Jul 18 11:58:31 CEST 2024

Screenshots / relevant information:

recon-all-log.txt (97.7 KB)

crash-20240718-115833-dc277359-autorecon1-42d5d326-c2fb-479f-ad51-15dbf9eb84d3.txt (14.2 KB)


Hi @Daniela,

In the future, please use the prepopulated software support template which prompts you for important information. You can see I reformatted your post this time.

I am going to guess this image is poor quality, resulting in recon-all using different commands that are not included in fmriprep container. You can try updating fmriprep. You can also run recon-all outside of fmriprep. But you might want to see if this image is quality enough to consider running at all.

Best,
Steven

Dear Steve,

Thank you for your reply. Next time, I’ll use the template.

This image is of very good quality; it is a 7T high-resolution fMRI with a voxel size of 1.2mm x 1.2mm. However, the way the protocol was programmed in the console makes the background grey instead of black (see image T1). We use a script to make it black before running the fmriprep (see image T1 modified, and the script attached).

I think the problem with the control point is that something is wrong with the intensity of the white matter (FsTutorial/ControlPoints_freeview - Free Surfer Wiki), possibly after changing the image.

Do you have any idea of what is happening or how can I correct this?

Best,
Daniela

02-denoiseAnat.txt (5.5 KB)