Reusing the GM-informed warping field to improve MTI registration#

Now that we have estimated a transformation between the template and the T2* data (based on the shape of the gray matter), we can use this warping field to improve a registration command from the previous MT registration tutorial step.

First, we return to the mt directory.

cd ../mt

Next, we run sct_register_multimodal to compute the transformation between the MT space and the PAM50 template space.

sct_register_multimodal -i "${SCT_DIR}/data/PAM50/template/PAM50_t2.nii.gz" \
                        -iseg "${SCT_DIR}/data/PAM50/template/PAM50_cord.nii.gz" \
                        -d mt1.nii.gz \
                        -dseg mt1_seg.nii.gz \
                        -param step=1,type=seg,algo=centermass:step=2,type=seg,algo=bsplinesyn,slicewise=1,iter=3 \
                        -m mask_mt1.nii.gz \
                        -initwarp ../t2s/warp_template2t2s.nii.gz \
                        -owarp warp_template2mt.nii.gz \
                        -qc ~/qc_singleSubj
Input arguments:
  • Since we are updating the previous tutorial, the input arguments are identical to ones from the registration step from that tutorial, except for one argument:

  • -initwarp : Here, we specify the T2* warping field generated in the previous step. We do this because information about the GM shape is enclosed in the transformation, which should producing registration results that are more accurate.

Output files/folders:
  • PAM50_t2_reg.nii.gz : The PAM50 template image, registered to the space of the MT1 image.

  • warp_template2mt.nii.gz : The warping field to transform the PAM50 template to the MT1 space.

https://raw.githubusercontent.com/spinalcordtoolbox/doc-figures/master/improving-registration-with-gm-seg/io-sct_register_multimodal-mt.png

This transformation can be used to warp the template to the MT space, which allows metrics to be extracted for specific vertebral levels, WM/GM tracts, and more.