brainweb-dl's Introduction

Brainweb-DL

Welcome to Brainweb-DL, a powerful Python toolkit for downloading and converting the Brainweb dataset.

Features

Effortless Dataset Management: Automatically download, cache, and format the Brainweb dataset with ease. Convert it to the convenient nifti format or numpy array hassle-free.
Multiple Image Generation: Generate high-quality T1, T2, T2*, and PD images directly from the Brainweb dataset. Perfect for testing purposes, although keep in mind that the values provided are approximate.

Available data

The Brainweb project kindly provides:

A normal brain phantom (named subject 0 afterwards), with T1, T2 and PD contrasts, with a variety of noise levels and intensity non-uniformities. As well as a anatomical model (in the form of either crisp or fuzzy segmentation of brain tissues, at a fixed resolution of 181x217x181 images).
The same for a multiple sclerosis brain phantom (named subject 1 afterwards).
A set of 20 normal brains (with ids equal to [4, 5, 6, 18, 20, 38, 41-54]) , with a T1 contrast (with 1mm resolution at (181, 217,181)), as well as the crisp and fuzzy segmentation of brain tissues (with a shape of (362, 434,362)) ¹.

This project provides a CLI and a Python API to download and convert theses data. On top of that, it can generate new contrasts (e.g. T2*) from the segmentations, and reshape the data to the desired resolution ².

Get Started

Data Location

The cached data directory follows this priority order:

User-specific argument (brainweb_dir in most functions)
BRAINWEB_DIR environment variable
~/.cache/brainweb folder

Python Script Usage

from brainweb_dl import get_mri 

data = get_mri(sub_id=44, contrast="T1")

The Brainweb dataset is downloaded and cached by default in the ~/.cache/brainweb folder.

Command Line Interface

brainweb-dl 44 --contrast=T1

For more information, see brainweb-dl --help.

Installation

Get up and running quickly!

pip install brainweb-dl

Development

Join our community and contribute to Brainweb-DL!

git clone [email protected]/paquiteau/brainweb-dl 
cd brainweb-dl
pip install -e .[dev,test,doc]

TODO List

Help us improve and shape the future of Brainweb-DL:

Add unit tests.
Implement fuzzy search and multiple subjects download in parallel.
Develop an interface to generate T1, T2, T2*, and PD images.
Enhance the search for the location of the Brainweb dataset (User > Environment Variable > Default Location).
Introduce an interface to download as BIDS format.

Acknowledgement

We extend our gratitude to the following for their contributions:

Casper De Clercq for the preliminary work and original idea. Check out his great work if you are interested in PET imaging and registration functionalities.
BrainWeb for providing this valuable dataset to the community.

🌟 If you like this work, don't forget to star it and share it 🌟

Note that the classification of tissue is not the same as for subject 0 and 1 ↩
This requires scipy to be installed. ↩

brainweb-dl's People

Contributors

Stargazers

Watchers

brainweb-dl's Issues

Issue with package import

Hi Pierre-Antoine, thank you for providing this nice brainweb extension! :)

I wanted to check out the fuzzy segmentation but I am struggling a bit with the import of the python package, as I am getting the following error:

Traceback (most recent call last): File "D:\PycharmProjects\project\stuff.py", line 5, in <module> import brainweb_dl File "C:\User\Anaconda3\envs\ptenv2\lib\site-packages\brainweb_dl\__init__.py", line 2, in <module> from .mri import get_mri File "C:\User\Anaconda3\envs\ptenv2\lib\site-packages\brainweb_dl\mri.py", line 11, in <module> from ._brainweb import ( File "C:\User\Anaconda3\envs\ptenv2\lib\site-packages\brainweb_dl\_brainweb.py", line 87, in <module> class BrainWebTissueMap(Enum): File "C:\User\Anaconda3\envs\ptenv2\lib\site-packages\brainweb_dl\_brainweb.py", line 88, in BrainWebTissueMap v1 = Path(str(files("brainweb_dl.data") / "brainweb1_tissues.csv")) File "C:\User\Anaconda3\envs\ptenv2\lib\importlib\resources.py", line 147, in files return _common.from_package(_get_package(package)) File "C:\User\Anaconda3\envs\ptenv2\lib\importlib\_common.py", line 14, in from_package return fallback_resources(package.__spec__) File "C:\User\Anaconda3\envs\ptenv2\lib\importlib\_common.py", line 18, in fallback_resources package_directory = pathlib.Path(spec.origin).parent File "C:\User\Anaconda3\envs\ptenv2\lib\pathlib.py", line 1082, in __new__ self = cls._from_parts(args, init=False) File "C:\User\Anaconda3\envs\ptenv2\lib\pathlib.py", line 707, in _from_parts drv, root, parts = self._parse_args(args) File "C:\User\Anaconda3\envs\ptenv2\lib\pathlib.py", line 691, in _parse_args a = os.fspath(a) TypeError: expected str, bytes or os.PathLike object, not NoneType

Do you have any idea how to fix this?

Recommend Projects

paquiteau / brainweb-dl Goto Github PK