Document data requirements for PeTu preprocessing by Copilot · Pull Request #23 · BrainLesion/PeTu

Copilot · 2025-11-08T19:56:53Z

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:

Multi-parametric MRI scans (T1c, T1n, T2w, T2f) are co-registered to T1c
Then brought into SRI-24 brain atlas space
Raw brain scans (without skull-stripping/defacing) should be used to preserve optic nerve structures
Recommendation to use the BrainLesion preprocessing package (with markdown links)
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>

Fixes [FEATURE] document data requirements #22

✨ 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>

neuronflow

@copilot please adress changes

neuronflow · 2025-11-08T20:04:47Z

README.md

+
+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).


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

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>

Copilot · 2025-11-08T20:06:52Z

@copilot please adress changes

Changes addressed in commit eee7cfc.

neuronflow · 2025-11-08T20:56:04Z

README.md

+
+**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.


use markdown syntac for hyperlinks

@copilot please adress comments

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

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

Copilot

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.

neuronflow

@copilot please implement changes based on comments

neuronflow · 2025-11-08T21:01:29Z

README.md

+
+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.


@copilot rather use this syntax

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

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

Copilot · 2025-11-08T21:03:44Z

@copilot please implement changes based on comments

Changes implemented in commit f516ee4.

Removed installation instructions for the preprocessing package.

Copilot

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.

MarcelRosier

one comment, otherwise LGTM

MarcelRosier · 2025-11-09T13:05:12Z

README.md

+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.


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

Initial plan

ceb22ae

Copilot AI assigned Copilot and neuronflow Nov 8, 2025

Copilot started work on behalf of neuronflow November 8, 2025 19:57 View session

Add Data Requirements section to README

2ba9702

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

Copilot finished work on behalf of neuronflow November 8, 2025 20:01

neuronflow requested changes Nov 8, 2025

View reviewed changes

Copilot started work on behalf of neuronflow November 8, 2025 20:05 View session

Clarify preprocessing: co-register to T1c then atlas space

eee7cfc

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

neuronflow requested a review from MarcelRosier November 8, 2025 20:07

Copilot AI requested a review from neuronflow November 8, 2025 20:08

Copilot finished work on behalf of neuronflow November 8, 2025 20:08

neuronflow requested changes Nov 8, 2025

View reviewed changes

Copilot started work on behalf of neuronflow November 8, 2025 20:56 View session

Use markdown syntax for hyperlinks

d823d99

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

Copilot AI requested a review from neuronflow November 8, 2025 20:59

Copilot finished work on behalf of neuronflow November 8, 2025 20:59

neuronflow requested a review from Copilot November 8, 2025 21:00

neuronflow marked this pull request as ready for review November 8, 2025 21:00

Copilot AI reviewed Nov 8, 2025

View reviewed changes

neuronflow requested changes Nov 8, 2025

View reviewed changes

Copilot started work on behalf of neuronflow November 8, 2025 21:02 View session

Use GitHub alert syntax for important note

f516ee4

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

Copilot AI requested a review from neuronflow November 8, 2025 21:04

Copilot finished work on behalf of neuronflow November 8, 2025 21:04

neuronflow requested a review from Copilot November 8, 2025 21:17

Update README to remove preprocessing package install

f1eba3a

Removed installation instructions for the preprocessing package.

neuronflow approved these changes Nov 8, 2025

View reviewed changes

Copilot AI reviewed Nov 8, 2025

View reviewed changes

neuronflow mentioned this pull request Nov 8, 2025

[QUESTION] Preprocessing pediatric brain data #21

Open

MarcelRosier approved these changes Nov 9, 2025

View reviewed changes

Update README.md

05ec702

neuronflow merged commit 71ae5f8 into main Nov 10, 2025
4 checks passed

neuronflow deleted the copilot/document-data-requirements branch November 10, 2025 08:47


		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).


		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.


		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.

Conversation

Copilot AI commented Nov 8, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!