Dear FSL experts,
My colleague and I are attempting to analyze DTI data in FSL. The first part ("DTIFIT Reconstruct Diffusion Tensor") worked just fine. I have outlined those steps below, just in case something we did inadvertently caused an issue in our later analysis. Our issue is with the Bedpostx application. This issue is described below, following the DTIFIT Reconstruct Diffusion Tensor steps we followed.
Step 1:
• In GUI, we selected "FDT Diffusion"
• We selected "Eddy Current Correction"
• Reference volume is 0
Step 2:
• We then made a brain mask for the DTI data
• We selected "Binary brain masking image"
• Fractional Intensity Threshold = 0.3
• We then unchecked "output brain-extracted image"
Step 3:
• This was followed by DTIFIT Reconstruct Diffusion Tensor
• We added the diffusion weighted data, BET Binary Mask, the bvec and bval files
• This ran without any difficulty
Step 4:
• We then viewed that data in FSL View without any difficulty. Everything worked as it should.
We now want to run bedpostx. We went back to FDT Diffusion, and then selected the Bedpostx option. We selected the folder with the participant's DTI data. The following are the contents of the folder:
dti_L2.nii.gz
dti_L3.nii.gz
dti_MD.nii.gz
dti_L1.nii.gz
dti_V2.nii.gz
dti_V3.nii.gz
dti_FA.nii.gz
dti_MO.nii.gz
dti_S0.nii.gz
dti_V1.nii.gz
nodif_brain_mask.nii.gz
data.nii.gz
data_ECC.ecclog
bvals - tried with and without text extension
bvecs - tried with and without text extension
The nii.gz files are from the DTIFIT Reconstruct Steps. Bvals and bvecs were included both with, and without the .txt extension (based on a possible lead we got from the mailing list - https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=FSL;60cb0eb.0812 ).
We then ran bedpostx and received a message "Child process ended abnormally." In OS X terminal, we see a message "bvecs not found" but this is odd because the bvecs file is in the directory.
Here is the content of the bvecs file:
0 -1 -1.41656e-05 1.00762e-06 0.0424009 -0.174908 -0.23229 -0.367522 -0.190198 0.116788 0.200486 0.495781 0.0141082 0.744492 0.760875 0.18088 0.679633 -0.777106 -0.924162 -0.468478 -0.881688 -0.690413 -0.239108 0.0577864 0.536766 0.991788 0.996763 0.872422 0.248702 -0.11831 -0.33761 -0.528597 -0.996853 0
0 -5.29348e-06 1 3.43559e-06 -0.114604 -0.200515 -0.162597 0.0260962 0.374378 0.833393 0.252684 0.134495 -0.628105 -0.147695 0.320392 0.924661 -0.422413 0.470692 -0.103601 -0.76738 -0.189303 0.706195 0.757082 0.983697 0.836063 -0.120693 0.0709022 0.478117 0.933555 0.991859 0.841492 0.840869 0.0549916 0
0 1.05469e-06 -3.44006e-06 1 -0.992506 -0.963951 -0.95896 -0.929649 -0.90756 -0.540201 -0.94655 -0.85797 -0.778001 -0.65109 -0.564286 -0.33509 -0.599722 -0.417799 -0.367683 -0.437784 -0.432193 -0.156902 -0.607992 0.170297 -0.113498 -0.0422999 -0.0378998 -0.101406 0.258112 -0.0471015 0.421794 0.116294 -0.0570967 0
We are running FSL on an iMac 10.8.2, with two memory slots (8 gb each). We are using FSL version 5.0.2.
What is preventing FSL from reading in the bvecs file? Also, would I be correct in assuming that this "child process ended abnormally" error would be the result of a "missing" bvecs file?
Thank you!
Best,
Victor DelBene
