Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Generalize ingression and normalization to handle T2w data #62

Open
tsalo opened this issue Aug 19, 2024 · 3 comments
Open

Generalize ingression and normalization to handle T2w data #62

tsalo opened this issue Aug 19, 2024 · 3 comments
Labels
enhancement New feature or request

Comments

@tsalo
Copy link
Member

tsalo commented Aug 19, 2024

          Note to self: open an issue that handles when there is a T2w available

Originally posted by @mattcieslak in #49 (comment)

@tsalo tsalo added the enhancement New feature or request label Aug 19, 2024
@tsalo
Copy link
Member Author

tsalo commented Aug 19, 2024

@mattcieslak what should QSIRecon do in the following scenarios? Which ones are already covered, or are impossible to reach?

  • No T1w or T2w
  • T1w only in native anatomical space
  • T2w only in native anatomical space
  • T1w only in MNI space
  • T2w only in MNI space
  • T1w and T2w in same native anatomical space
  • T1w and T2w in different anatomical spaces
  • T1w and T2w in same MNI space

@mattcieslak
Copy link
Contributor

The two key pieces of anatomical data that qsirecon uses are

  1. Whether there is a high-res anatomical in space-T1w
  2. Whether there are transforms to and from MNI2009cNlinAsym

Here's what currently happens:

  • No T1w or T2w

If will try to load in a T1w from the --freesurfer-input and will register that to the dwi reference. If there's nothing, then it makes any workflows that require an anatomical image raise an error.

  • T1w only in native anatomical space

The resampled T1w in template space isn't used for anything in qsirecon.

  • T2w only in native anatomical space

I think it results in an error because t1w is hardcoded into the ingress interfaces.

  • T1w only in MNI space
    I've never seen this happen, but this would result in the same behavior as if there is no t1w space anat.
  • T2w only in MNI space

Same as ^

  • T1w and T2w in same native anatomical space

Ignores T2w and uses T1w in everything

  • T1w and T2w in different anatomical spaces

As long as there is a space-T1w T1w image and a pair of transforms to MNI, things will function normally

  • T1w and T2w in same MNI space

Both will be ignored.

@tsalo
Copy link
Member Author

tsalo commented Aug 19, 2024

Regarding the T2w-related scenarios, what should happen (esp. with HBCD data)? The main one is when T1w and T2w are both present. Should QSIRecon default to using the T2w as the main anatomical modality when both are present? I assume that should depend on if infant mode is enabled. E.g., T1w as primary for adults, T2w as primary for infants. WDYT?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

2 participants