Hi Mark,
Thanks so much for your reply. I have checked the registrations and they all seem fine. It is strange because, although I have got 2 left cerebellum failures like this (with the no interior voxels error) and 10 right cerebellum, in all cases the opposite cerebellum segmentation worked eg. if the left failed the right segmentation worked ok and vice versa. I am not 100% sure what information you want so here is an example of what comes up when I do fslinfo on one of the input whole head scans:
data_type INT16
dim1 164
dim2 240
dim3 240
dim4 1
datatype 4
pixdim1 1.1000009775
pixdim2 1.0000000000
pixdim3 1.0000000000
pixdim4 0.0000000000
cal_max 1000.0000
cal_min 0.0000
file_type NIFTI-1+
Some of the scans have different dim1, dim2 and dim3 values because they were obtained on different scanners.
The script I have been using is slightly different from the norm because a colleague who started segmenting caudates using FSL found that she was getting a lot of registration failures. She recommended doing the BET script to extract the brain and use the betted brain for the registration. Then, the transformation matrix from the betted brain registration is applied to the original whole head scan in the FIRST segmentation stage. Here is an example of the script I am using:
run_first -i wholeheadscan -t bettedtransformationmatrix-to-std-sub_tmp_cort_stage2.mat -n 40 -o scannumber_L_cereb_intref_cort_n40 -v -intref ${FSLDIR}/data/first/models_336_bin/05mm/L_Puta_05mm.bmv -m ${FSLDIR}/data/first/models_336_bin/intref_puta/L_Cereb.bmv
I have copied an example of what happens when I run the script at the end of this message in case that helps. Sorry this is such a long message, but I think it best that you have as much information as possible! I am sure that the problem will turn out to be something simple, but it is great to have help working through sensible options.
Thank you,
Clare
read model
done reading model
setting up shape/appearance model
The shape has 642 vertices.
The model was constructed from 336 training subjects.
336 modes of variation are retained.
create shapeModel
done creating shapeModel
read model
done reading model
setting up shape/appearance model
The shape has 642 vertices.
The model was constructed from 336 training subjects.
336 modes of variation are retained.
create shapeModel
done creating shapeModel
reading image 20067-013-1
normalize intensity...
Reading transformation matrix 20067-013-1-to-std-sub_tmp_cort_stage2.mat
0.990658 0.00540601 -0.00246533 0.497929
-0.00599331 0.988266 -0.00182927 1.72579
0.00224005 0.00967193 0.97641 0.399656
0 0 0 1
1.00939 -0.00554641 0.00253821
0.00611702 1.01182 0.00191106
-0.00237631 -0.01001 1.02414
NEw done imodes transform
done registering model
mode 157.046
found mode 157.046
vars:
-0.665811 0.974802 -1.24049 0 -0.995397 -0.476898 -1.02865 -0.805348 0.496544 -0.382277
vars:
3.66446 0.974802 -1.24049 -3.31629 -3.47849 5.51211 -10.0764 3.01544 0.496544 -1.58938
vars:
2.64391 1.83909 0.357607 -3.36717 -7.08195 3.25283 -10.2576 1.29871 0.200706 -2.9231
vars:
4.80036 0.376847 0.357607 -3.05923 -9.33776 5.41102 -5.67457 1.29871 2.98748 -2.9231
vars:
5.25619 1.91634 0.385879 -3.05923 -11.3662 3.91984 -5.67457 1.29871 2.98748 -5.18784
vars:
7.23852 1.91634 1.14347 -3.05923 -10.7433 1.74672 -3.3057 3.61577 5.49094 -0.0171847
vars:
7.23852 1.58512 0.305583 -3.05923 -12.2404 -0.169016 -6.51417 3.18212 7.08267 1.16017
vars:
7.23852 2.25733 0.305583 -3.76332 -14.9787 -0.169016 -5.4902 2.16117 7.6049 -0.881496
vars:
7.23852 1.97441 0.0316077 -3.98942 -15.1027 -0.169016 -5.88181 2.31508 7.6049 -0.881496
vars:
7.23852 1.78081 0.0316077 -4.17977 -15.2266 -0.574529 -5.88181 2.33688 7.36112 -0.683371
vars:
7.23852 1.78081 0.0316077 -4.17977 -15.2266 -0.574529 -5.88181 2.33688 6.78165 -0.683371
vars:
7.23852 1.78081 0.0316077 -4.17977 -15.7636 -0.574529 -5.88181 2.33688 7.07682 -0.683371
vars:
7.23852 1.48801 0.0316077 -3.88874 -15.7636 -0.574529 -5.88181 2.48103 6.67038 -0.683371
vars:
7.23852 1.48801 0.0316077 -3.88874 -15.7636 -0.574529 -5.5012 2.48103 6.29976 -0.507553
vars:
7.23852 1.48801 0.0316077 -3.88874 -15.7636 -0.574529 -5.5012 2.48103 6.29976 -0.507553
vars:
7.23852 1.34448 0.0316077 -3.88874 -15.9724 -0.574529 -5.57673 2.56491 6.29976 -0.576234
vars:
7.23852 1.34448 0.0316077 -3.88874 -15.9724 -0.63252 -5.4218 2.56491 6.0668 -0.576234
vars:
7.23852 1.34448 0.0316077 -3.88874 -15.9724 -0.63252 -5.4218 2.56491 6.0668 -0.576234
WARNING: NO INTERIOR VOXELS TO ESTIMATE MODE
|