search

sct-pipeline / spine-park

Pipeline for multicontrast analysis in PD patients
MIT License
0 stars 0 forks source link

Issue with the segmentation of dwi images `sub-DEV205Sujet09` #38

Closed Kaonashi22 closed 5 months ago

Kaonashi22 commented 5 months ago

For one subject, the aorta has been segmented on the dwi images, chunk 3. image

Here is the position of the WM template after registration ot the template. image

I was using this SCT version when I had the issue: git version 2.25.1 And this script version: image

@jcohenadad , I'll send you the data by email

jcohenadad commented 5 months ago

I've updated the title to include the subject name.

I've ran the processing using script version 46499f4fa7e7ad2bf6663763736af9863461cc58, SCT git-master-9e1f65b86902cb237d5f79058a4623eed2d2c8eb and these derivatives: sub-DEV205Sujet09.zip

I did not observe the issue reported here: https://github.com/sct-pipeline/spine-park/issues/38#issue-2364342788. Here are my log files and QC reports:

@Kaonashi22 can you please try with the derivatives I've attached, and upload the generated QC report and log file please.

Kaonashi22 commented 5 months ago

Sorry, I forgot you send you my derivatives for this subject: sub-DEV205Sujet09.zip

I am rerunning the analyses with your derivatives ; I'll let you know what I find.

Kaonashi22 commented 5 months ago

Using the script version https://github.com/sct-pipeline/spine-park/commit/46499f4fa7e7ad2bf6663763736af9863461cc58 and SCT git-master-9e1f65b86902cb237d5f79058a4623eed2d2c8eb with your derivatives, I didn't reproduce the issue either.

batch_processing_sub-DEV205Sujet09.log qc.zip

Kaonashi22 commented 5 months ago

The spinal cord segmentation is also slighting different between the two runs : -"seg_1" : first segmentation with the issue -"seg_2" : last segmentation without the issue

seg_2_sub-DEV205Sujet09_chunk-3_DWI_dwi_mean_seg.nii.gz [Uploading sub-DEV205Sujet09_chunk-3_DWI_dwi_mean.nii.gz…]() seg_1_sub-DEV205Sujet09_chunk-3_DWI_dwi_mean_seg.nii.gz

Kaonashi22 commented 5 months ago

Do you think I can use my derivatives from my previous run (two months ago) when I run the pipeline with the updated script and SCT version? I did an extensive manual correction on the segmentation masks. Or would you recommend redoing everything from scratch?

jcohenadad commented 5 months ago

Do you think I can use my derivatives from my previous run (two months ago) when I run the pipeline with the updated script and SCT version? I did an extensive manual correction on the segmentation masks.

Yes, absolutely. You can reuse the segmentations you've already corrected if you are happy with them.

jcohenadad commented 5 months ago

I assume we can close this issue now