Different result in computing CSA

SCT
1

Dear all
When I compute cross-sectional area (CSA) of spinal cord and average it across levels C3 and C4

(venv_sct) forest@DESKTOP-DM8SQ21:~/sct_course_london20/single_subject/data/t2$ sct_process_segmentation -i t2_seg.nii.gz -vert 3:4 -o csa_c3c4.csv

--
Spinal Cord Toolbox (git-HEAD-95e66208c9d6937fbe00e86db6176e4001196e56)

sct_process_segmentation -i t2_seg.nii.gz -vert 3:4 -o csa_c3c4.csv
--

Compute shape analysis: 100%|##############| 241/241 [00:00<00:00, 278.26iter/s]

Done! To view results, type:
xdg-open /home/forest/sct_course_london20/single_subject/data/t2/csa_c3c4.csv

This is my result

image
image1893×80 9.87 KB

It’s a little different from your result

image
image958×692 165 KB

Looking forward to your reply!

2

Hi @chang

Thank you for reaching out. You made an excellent observation! Indeed, results are slightly different because they were produced with an older version of SCT. Since then, we have fixed a few bugs and improved the overall software, so some changes are sometimes expected.

In order to let users (and developers) know if a new release has an impact on the main quantitative functionalities of SCT, for each new release, we run a typical quantitative analysis on two different OSs, and report the results in the release notes. Example here: https://github.com/neuropoly/spinalcordtoolbox/releases/tag/5.2.0 (click on “Results of batch_processing.sh”)

Best
Julien