Question about alias/symlink files in FreeSurfer "surf" directory after t1-freesurfer

[Muhammad Orhan]

Hi Clinica Team,

During t1-freesurfer processing, I noticed that for some subjects a few files inside the surf folder appear as aliases (symlinks) rather than original files, as shown in the attached screenshot. This does not happen for all subjects.

Is this expected behaviour in FreeSurfer, or does it indicate a problem with the reconstruction?
More importantly, could these symlink files affect the PET-Surface pipeline later on, or is Clinica designed to work with them without issues?

Setup: MAC-M3, Clinica 0.94 / 10.1, FreeSurfer: (freesurfer-macOS-darwin_x86_64-7.4.1)

Thanks in advance for your help!

Best regards,

Muhammad

---------------------------

[Alice Joubert]

Hello Muhammad,

Indeed this could indicate a process completion issue and I don't think pet-surface is able to handle symlinks. I will try to test this case next week if I have time. Does this represent a lot of your subjects? 

Best,

Alice

--
You received this message because you are subscribed to the Google Groups "Clinica" group.
To unsubscribe from this group and stop receiving emails from it, send an email to clinica-user...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/clinica-user/CAN5GZA0hXKsttvefdkeZ-bw6L9c8ST9H1119E9%2BGKU-p0XN8Ww%40mail.gmail.com.

[Muhammad Orhan]

Hello! A. Joubert,
Thanks.. that would be great if you test this.
I faced this issue with around 400 subjects from a cohort of 1500.

To view this discussion visit https://groups.google.com/d/msgid/clinica-user/CAP_epFvkLBdgqCaMc_fKSvWP5Oa1xRdZ%3DwAt6AeV_OUoCNu6eA%40mail.gmail.com.

[Alice Joubert]

Hello Muhammad,

Sorry for the delay. I tried to reproduce the bug with subject `003_S_1122`, first session. I could not even create a bids for the image as it does not pass our converter QC check. How did you obtain your nifti for it ?

Best,

Alice

[Muhammad Orhan]

Hi Alice,

Thanks for your response. that’s surprising actually. I generated the nifti using Clinica v0.9.4, without any manual modification.

Best,

Muhammad

To view this discussion visit https://groups.google.com/d/msgid/clinica-user/b3af96ca-8584-444b-9b78-a91ae11ad45cn%40googlegroups.com.

[Alice Joubert]

Hello Muhammad,

Okay, I will try again then. How did you select your images on the loni ADNI website ? (preprocessed, processed, etc... if you know what I am referring to)

Best,

Alice

[Muhammad Orhan]

Yes, I used ADNI converter to download data.. i selected all baseline scans.

———————

To view this discussion visit https://groups.google.com/d/msgid/clinica-user/8a0f9479-8eea-4879-9cfd-2b0a4b97bbc0n%40googlegroups.com.

[Alice Joubert]

I cannot seem to reproduce with this subject even when taking every data available. Do you have any other subjects that did not work for you ?

[Muhammad Orhan]

Dear Alice,
I have attached a list of subject which runs through same problem.
Please have a look.

Thanks. 

Regards

To view this discussion visit https://groups.google.com/d/msgid/clinica-user/2bcbddc5-2229-4420-9315-75075a713af7n%40googlegroups.com.

[Alice Joubert]

Hello, 

I have been able to reproduce now. I am unsure as to why it does this and will try to investigate further but I want to work on our next release first. Did you try running the PET-Surface just to see what it would do ?

Best,

Alice