Skip to content

Conversation

Copy link
Contributor

Copilot AI commented Nov 8, 2025

Document Data Requirements

  • Add "Data Requirements" section to README.md after the Features section
  • Include information about:
    • PeTu training on Kispi data with optic glioma affecting the optic nerve
    • Input should be raw brains without defacing or skull-stripping
    • Note about trying with BET or defaced brain scans
    • SRI-24 atlas usage
    • Recommendation to use BrainLesion preprocessing package
  • Clarify preprocessing steps: scans should be co-registered to T1c, then brought into SRI-24 atlas space
  • Use markdown syntax for hyperlinks instead of HTML tags
  • Use GitHub alert syntax for important note (> [!IMPORTANT])
  • Address all reviewer feedback

Summary

Successfully documented data requirements with clear preprocessing steps:

  1. Multi-parametric MRI scans (T1c, T1n, T2w, T2f) are co-registered to T1c
  2. Then brought into SRI-24 brain atlas space
  3. Raw brain scans (without skull-stripping/defacing) should be used to preserve optic nerve structures
  4. Recommendation to use the BrainLesion preprocessing package (with markdown links)
  5. Important note formatted using GitHub alert syntax for better visibility

This addresses issue #21 and clarifies the exact preprocessing procedure for pediatric brain data.

Original prompt

This section details on the original issue you should resolve

<issue_title>[FEATURE] document data requirements</issue_title>
<issue_description>We need something similar to this, like we have for the BraTS orchestrator. However, here we do not employ BET

## Data  Requirements
BraTS challenge algorithms require preprocessed brain scans.  
This preprocessing typically includes co-registration, brain extraction, and registration to a challenge-specific brain atlas (template).  
Please refer to each challenge’s documentation for details on which template to use.  

For convenience, we provide challenge-specific wrappers around our [preprocessing package](https://github.com/BrainLesion/preprocessing), part of the [BrainLesion Suite](https://github.com/BrainLesion).  
You can install the package with the preprocessing extra using:  


```bash
pip install brats[preprocessing]

The modular architecture of the BraTS Orchestrator facilitates employing your own preprocessing routines.
Alternatively, the preprocessing package can be used independently to design custom preprocessing pipelines tailored to your specific needs.


<agent_instructions>Please document the data requirements here is what we wrote for another package..however, PETU is trained on Kispi data...because we have optic glioma affecting the optic nerve the input should be raw brains without defacing or skullstripping (even though it is worth trying with BET or defaced brain scans)..

We used the SRI-24 atlas
## Data Requirements

GlioMODA is trained on [BraTS](https://github.com/BraTS) (Brain Tumor Segmentation) preprocessed images.
This preprocessing typically includes co-registration, skull-stripping (brain extraction), and registration to the SRI-24 brain atlas (template).

We recommend using the [preprocessing package](https://github.com/BrainLesion/preprocessing), part of the [BrainLesion Suite](https://github.com/BrainLesion), to design custom preprocessing pipelines tailored to your specific needs.
You can install the package with:

```bash
pip install brainles-preprocessing
```</agent_instructions>

## Comments on the Issue (you are @copilot in this section)

<comments>
<comment_new><author>@neuronflow</author><body>
this is necessary as we see in https://github.com/BrainLesion/PeTu/issues/21</body></comment_new>
</comments>


✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

Co-authored-by: neuronflow <7048826+neuronflow@users.noreply.github.com>
Copilot AI changed the title [WIP] Add documentation for data requirements Document data requirements for PeTu preprocessing Nov 8, 2025
Copilot AI requested a review from neuronflow November 8, 2025 20:01
Copy link
Contributor

@neuronflow neuronflow left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@copilot please adress changes

README.md Outdated

PeTu is trained on pediatric brain MRI data from the Children's Hospital Zurich (Kispi), including cases with optic glioma affecting the optic nerve.

The model expects co-registered multi-parametric MRI scans (T1c, T1n, T2w, T2f) that have been registered to the SRI-24 brain atlas (template).
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

scans should be co-registered to T1c and then brought into atlas space.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Updated the line to clarify that scans should be co-registered to T1c and then brought into SRI-24 atlas space (eee7cfc).

Co-authored-by: neuronflow <7048826+neuronflow@users.noreply.github.com>
Copy link
Contributor Author

Copilot AI commented Nov 8, 2025

@copilot please adress changes

Changes addressed in commit eee7cfc.

README.md Outdated

**Important:** Since PeTu handles optic gliomas affecting the optic nerve, input data should be **raw brain scans without defacing or skull-stripping** to preserve critical anatomical structures. However, it may be worth experimenting with skull-stripped (BET) or defaced brain scans depending on your specific use case.

We recommend using the <a href="https://github.com/BrainLesion/preprocessing">preprocessing package</a>, part of the <a href="https://github.com/BrainLesion">BrainLesion Suite</a>, to design custom preprocessing pipelines tailored to your specific needs.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

use markdown syntac for hyperlinks

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@copilot please adress comments

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Converted HTML <a> tags to markdown syntax [text](url) (d823d99).

Co-authored-by: neuronflow <7048826+neuronflow@users.noreply.github.com>
Copilot AI requested a review from neuronflow November 8, 2025 20:59
@neuronflow neuronflow requested a review from Copilot November 8, 2025 21:00
@neuronflow neuronflow marked this pull request as ready for review November 8, 2025 21:00
Copy link

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull Request Overview

This PR adds a new "Data Requirements" section to the README, providing essential information about the training data, expected input formats, and preprocessing recommendations for users of the PeTu pediatric brain tumor segmentation tool.

  • Added comprehensive data requirements section explaining training data source, expected MRI modalities, and preprocessing needs
  • Included important guidance about using raw brain scans for optic glioma cases
  • Added reference to the BrainLesion preprocessing package with installation instructions

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copy link
Contributor

@neuronflow neuronflow left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@copilot please implement changes based on comments

README.md Outdated

The model expects multi-parametric MRI scans (T1c, T1n, T2w, T2f) that are co-registered to T1c and then brought into SRI-24 brain atlas space.

**Important:** Since PeTu handles optic gliomas affecting the optic nerve, input data should be **raw brain scans without defacing or skull-stripping** to preserve critical anatomical structures. However, it may be worth experimenting with skull-stripped (BET) or defaced brain scans depending on your specific use case.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@copilot rather use this syntax

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Changed to use GitHub alert syntax > [!IMPORTANT] instead of bold text (f516ee4).

Co-authored-by: neuronflow <7048826+neuronflow@users.noreply.github.com>
Copy link
Contributor Author

Copilot AI commented Nov 8, 2025

@copilot please implement changes based on comments

Changes implemented in commit f516ee4.

Removed installation instructions for the preprocessing package.
Copy link

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull Request Overview

Copilot reviewed 1 out of 1 changed files in this pull request and generated no new comments.


💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copy link
Contributor

@MarcelRosier MarcelRosier left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

one comment, otherwise LGTM

README.md Outdated
The model expects multi-parametric MRI scans (T1c, T1n, T2w, T2f) that are co-registered to T1c and then brought into SRI-24 brain atlas space.

> [!IMPORTANT]
>Since PeTu handles optic gliomas affecting the optic nerve, input data should be **raw brain scans without defacing or skull-stripping** to preserve critical anatomical structures. However, it may be worth experimenting with skull-stripped (BET) or defaced brain scans depending on your specific use case.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

raw might sound a bit misleading since we do expect registration?

@neuronflow neuronflow merged commit 71ae5f8 into main Nov 10, 2025
4 checks passed
@neuronflow neuronflow deleted the copilot/document-data-requirements branch November 10, 2025 08:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[FEATURE] document data requirements

3 participants