PreFreeSurferPipelineBatch.sh didn't average multiple T1w images correctly.

[AbestSG]

Hello,

I ran the PreFreeSurferPipelineBatch.sh on four T1w images but the T1w.nii.gz didn't look right in the output folder. Could you please help me with the issue?

Thanks,
Si

[coalsont]

What do the T1w/T1w1_GradientDistortionUnwarp/T1w1 files look like individually (which is after only the fslreorient2std step)? If the scanner has lied about the patient's approximate head orientation, then I wouldn't be surprised that the subsequent registration failed.

[AbestSG]

What do the T1w/T1w1_GradientDistortionUnwarp/T1w1 files look like individually (which is after only the fslreorient2std step)? If the scanner has lied about the patient's approximate head orientation, then I wouldn't be surprised that the subsequent registration failed.

Hello,

Thanks for the reply. It starts looking weird in the T1w2_gdc_to_im1_linmask.nii.gz however another T1w3_gdc_to_im1_linmask.nii.gz looks normal.

Thanks,
Si

[glasserm]

What is the age of the participant and what do the original images look like?

[AbestSG]

The original image looks normal. Age is 24 years old.

I just tested the commands in the AnatomicalAverage.sh script. I found if I remove '-refweight ${im1}_roi_linmask' from the following command in the script and re-run the pipeline, the issue is gone and the T1w2_gdc_to_im1_linmask.nii.gz and T1w.nii.gz look good.

#$FSLDIR/bin/flirt -in ${im2}_roi -init ${im2}_to_im1.mat -omat ${im2}_to_im1_linmask.mat -out ${im2}_to_im1_linmask -ref ${im1}_roi -refweight ${im1}_roi_linmask -nosearch

[glasserm]

Why are you mixing images with and without receive field correction (pre-scan normalize)? I would start by not doing that before you start editing the HCP Pipelines.