4.0.2 MTR value error


#1

Hello everyone,

I recently upgrade to 4.0.2 version. I found an error on calculation MTR.

I use the same commands and the same cases calculating MTR; however, 4.0.2 gave error value different from 4.0.0 beta version.

Command:
sct_deepseg_sc -i mt1.nii -c t2 -qc ~/qc_singleSubj

sct_create_mask -i mt1.nii -p centerline,mt1_seg.nii -size 35mm -f cylinder -o mask_mt1.nii

sct_register_multimodal -i mt0.nii -d mt1.nii -dseg mt1_seg.nii -param step=1,type=im,algo=slicereg,metric=CC -m mask_mt1.nii -x spline -qc ~/qc_singleSubj

sct_compute_mtr -mt0 mt0_reg.nii -mt1 mt1.nii

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 -dseg mt1_seg.nii -param step=1,type=seg,algo=centermass:step=2,type=seg,algo=bsplinesyn,slicewise=1,iter=3 -m mask_mt1.nii -initwarp …/t2/warp_template2anat.nii.gz -qc ~/qc_singleSubj

sct_label_utils -i mt1_seg.nii -create-seg -1,4 -o label_c3c4.nii.gz

sct_register_to_template -i mt1.nii -s mt1_seg.nii -ldisc label_c3c4.nii.gz -ref subject -param step=1,type=seg,algo=centermassrot:step=2,type=seg,algo=bsplinesyn,slicewise=1

mv warp_PAM50_t22mt1.nii.gz warp_template2mt.nii.gz

sct_warp_template -d mt1.nii -w warp_template2mt.nii.gz -qc ~/qc_singleSubj

sct_extract_metric -i mtr.nii -l 50 -vert 3:4 -o mtr_in_sc3_4.csv

Results:
4.0.0beta5
2019-10-18 21:22:48 4.0.0-beta.5 /Users/ohlaboratory/Desktop/ris/mt/mtr.nii 2:12 3:4 spinal cord 1915.1919487633800 33.395869749980800 5.913861825314080

4.0.2
2019-10-22 15:55:02 4.0.2 /Users/ohlaboratory/Desktop/ris0203002/mt/mtr.nii 2:12 3:4 spinal cord 1902.0600379818200 178398311743488.0 0.0

Another issue mtr.nii cannot show proper on FSLviewer even though I have adjust the window width/level.

Could you please let me know how to fix these issues?

Thank you again.


#2

Hi @Suradech_Suthiphosuw,

would it be possible to send me the files mt0_reg.nii and mt1.nii so I can identify the cause of the problem?

Cheers,
Julien


#3

Hello Julien,

Thank you for your prompt reply. I sent the files to your email. Let me know …


#4

Hi @Suradech_Suthiphosuw

I was able to reproduce the error. I will implement a fix very soon. Thank you for reporting this bug.


#5

Thank you so much again.


#6

Hi @Suradech_Suthiphosuw

Just to let you know that the bug has been fixed (more details here).

You can already work on the fixed version by using the github installation, or wait for the next release (4.1).

Best,
Julien


#7

It works now. Thank you so much!