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

Add explanations for 1st and 2nd level for glm #2599

Open
NicolasGensollen opened this issue Nov 12, 2020 · 6 comments · May be fixed by #4287
Open

Add explanations for 1st and 2nd level for glm #2599

NicolasGensollen opened this issue Nov 12, 2020 · 6 comments · May be fixed by #4287
Labels
Documentation for documentation related questions or requests GLM Issues/PRs related to the nilearn.glm module. Good first issue Good for newcomers. Equivalent to "very low" effort.

Comments

@NicolasGensollen
Copy link
Member

Add a few lines somewhere in the doc to explain what are glm first and second levels which might not be obvious for researchers from other fields.
Note: We could also put a link to these explanations in the html reports.

@NicolasGensollen NicolasGensollen added the Documentation for documentation related questions or requests label Nov 12, 2020
@NicolasGensollen NicolasGensollen added the Good first issue Good for newcomers. Equivalent to "very low" effort. label Mar 18, 2021
@NicolasGensollen NicolasGensollen added this to Stalled PRs in Nilearn Dev Days 2021 via automation Mar 18, 2021
@NicolasGensollen NicolasGensollen moved this from Stalled PRs to Documentation in Nilearn Dev Days 2021 Mar 18, 2021
@Remi-Gau Remi-Gau removed this from Documentation in Nilearn Dev Days 2021 Apr 27, 2023
@Remi-Gau Remi-Gau added the GLM Issues/PRs related to the nilearn.glm module. label Aug 28, 2023
@Remi-Gau Remi-Gau added this to the release 0.11.0 milestone Oct 7, 2023
@tpremrud
Copy link

@NicolasGensollen Thank you for opening the issue!
Are these (i.e., https://nilearn.github.io/dev/glm/first_level_model.html, https://nilearn.github.io/dev/glm/second_level_model.html) the documentation links to first level and second level GLMs, respectively?

It seems there are some information in those pages already. With that, is this issue still open as I would love to contribute!

Thank you!!!

@bthirion
Copy link
Member

You're welcome ! Please review carefully the existing doc and identify missing stuff.

@Remi-Gau
Copy link
Collaborator

Remi-Gau commented Feb 1, 2024

@tpremrud

From @NicolasGensollen original message I am not sure exactly what information is actually missing.

I do find the summaries for those pages good enough. But then again I have been dealing fMRI GLMs for a while now, so I think I may not even realize what implicit information may be missing and that I am filling in. So please, have a look and don't hesitate to point out what is missing in your opinion.

Random thoughts...

Maybe the issue was referring to what the main classes for first and second level actually do ?

Or maybe we should "explain" things in terms of:

  • "run level GLM"
  • "subject level GLM"
  • "group level GLM"
    as different software may mean different things for First and Second level.
    Also to align better with the Nodes of the bids stats model formalization.

@tpremrud
Copy link

tpremrud commented Feb 2, 2024

@Remi-Gau Thank you for pointing out that it seems complete already! I could definitely fill in the missing pieces in terms of implicit information, as I am not an expert in fMRI GLM.

With that, I think both random thoughts could be taken care of, but I am struggling to understand what you meant by 'what the main classes for first and second level actually do'. Is it expanding more on the class definition given as 'Implement the General Linear Model for single run fMRI data.' for the first level? Or are you talking about expanding the function definition more, which might not be clear enough (e.g., fit method)?

For the second random thought, where do you think we should address that diversity in meanings? I think the GitHub pages would be a great place to update, but I'm not sure if I'm missing anything.

Thank you so much!
PS: This is my first time ever contributing to a GitHub repo issue, so I am excited to contribute but maybe a bit too nervous to see what is going on; hence, I appreciate any help! :)

@Remi-Gau
Copy link
Collaborator

Remi-Gau commented Feb 5, 2024

I am struggling to understand what you meant by 'what the main classes for first and second level actually do'. Is it expanding more on the class definition given as 'Implement the General Linear Model for single run fMRI data.' for the first level?

I was more thinking about this one

For the second random thought, where do you think we should address that diversity in meanings? I think the GitHub pages would be a great place to update

Maybe adding some explanation in this page of the doc?

Corresponding to this file: https://github.com/nilearn/nilearn/blob/main/doc/modules/glm.rst

Or in the "index" page of the GLM section of the user guide?

Corresponding to this file: https://github.com/nilearn/nilearn/blob/main/doc/glm/index.rst

Thank you so much! PS: This is my first time ever contributing to a GitHub repo issue, so I am excited to contribute but maybe a bit too nervous to see what is going on; hence, I appreciate any help! :)

Woohoo!!! Exciting!!! Though thinking back on my first ever contribution, I can relate to the being nervous part.

If things are not clear: don't hesitate to ask.

Before you jump in, make sure you have look at the "how to contribute" section of the doc.

Especially this part: https://nilearn.github.io/stable/development.html#setting-up-your-environment

You could also come to the drop in hours if you need guidance on the content of your contributions or how to go about it.

#3211

@tpremrud
Copy link

After some guidance from @Remi-Gau, I have done a PR on comparing the meaning of different GLM (first, second, and third-level models) for FSL and SPM, with the possible extension of extending to comparing with those of AFNI.

Feedbacks are truly appreciated! :D

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Documentation for documentation related questions or requests GLM Issues/PRs related to the nilearn.glm module. Good first issue Good for newcomers. Equivalent to "very low" effort.
Projects
None yet
Development

Successfully merging a pull request may close this issue.

4 participants