Print

---

Dear all!

I am trying to register JHU atlas to FA native space and then create ROIs to extract mean values. But I am almost a complete noobie, it is the first time I use flirt fnirt etc... and I have a few doubts that I will really appreciate if you could help. 

So far, this is what I have done (I don't know if it will crash, it's still working on the first subject): 

bet qsm_${i}_T1.nii.gz ${i}_betT1.nii.gz

echo "flirt ${i}FA to T1" 
	flirt -ref ${i}_betT1.nii.gz -in ${i}FA.nii.gz -dof 6 -omat ${i}FA_to_T1_rig.mat -out ${i}_FA2T1.nii.gz

echo "flirt ${i}T1 to MNI" 

flirt -ref /usr/share/data/fsl-mni152-templates/MNI152_T1_1mm_brain.nii.gz -dof 12 -in ${i}_betT1.nii.gz -omat ${i}_aff_T12mni.mat -out ${i}_affT12MNI.nii.gz

echo "FNIRT ${i}T1 to MNI" 

fnirt --ref=/usr/share/data/fsl-mni152-templates/MNI152_T1_1mm_brain.nii.gz --in=${i}_betT1.nii.gz --aff=${i}_aff_T12mni.mat --cout=${i}_non_lin_T12mni.nii.gz

##concatenate 3 transformations

convert_xfm -omat ${i}_FA2mni.mat -concat ${i}_non_lin_T12mni.nii.gz ${i}_aff_T12mni.mat ${i}b0_to_T1_rig.mat

##inverse transformation 
convert_xfm -omat ${i}_MNI2FA.mat -inverse ${i}_FA2mni.mat

echo "applywarp to JHU for ${i}"

applywarp --ref=${i}FA.nii.gz --in=/usr/share/data/jhu-dti-whitematter-atlas/JHU/JHU-ICBM-tracts-maxprob-thr25-1mm.nii.gz --out=${i}_JHU2FA.nii.gz --coef=${i}_MNI2FA.mat

my doubts are:

1) I've used FA images directly because as I understood eddy does align the volumes to b0, should I use b0 though?

2) is concatenation between the transformations a correct step?

3) instead of using invwarp , I Inverted the warp and then I used applywarp, is this correct?

4) fnirt is running now. The terminal output is saying: "Warning, Jacobian not within prescribed range. Prescription is 0.01 -- 100 and obtained range is -10.3426 -- 85.9382" What does it mean :(?

5) I don't know anything about atlases,,. should I have used another in the case of MNI? Should I have used another in the case of he JHU atlas? 

6) In case this works... now I would have to extract masks  

THANK YOU SOOOOO MUCH FOR READING

I hope this helps like other messages from the lsit have helped me!