/atlas-alignment

Blue Brain multi-modal registration and alignment toolbox

Primary LanguagePythonGNU Lesser General Public License v3.0LGPL-3.0

Atlas Alignment

Latest Release Latest release
License License
Data Data
Build Status Build status
Code Style Black Isort Pydocstyle Pydocstyle
Python Versions Python Versions

Atlas Alignment is a toolbox to perform multimodal image registration. It includes both traditional and supervised deep learning models.

This project originated from the Blue Brain Project efforts on aligning mouse brain atlases obtained with ISH gene expression and Nissl stains.

Official documentation

All details related to installation and logic are described in the official documentation.

Installation

Installation Requirements

Some of the functionalities of atlalign depend on the TensorFlow implementation of the Learned Perceptual Image Patch Similarity (LPIPS). Unfortunately, the package is not available on PyPI and must be installed manually as follows.

pip install git+http://github.com/alexlee-gk/lpips-tensorflow.git#egg=lpips_tf

You can now move on to installing the actual atlalign package!

Installation from PyPI

The atlalign package can be easily installed from PyPI.

pip install atlalign

Installation from source

As an alternative to installing from PyPI, if you want to try the latest version you can also install from source.

pip install git+https://github.com/BlueBrain/atlas_alignment#egg=atlalign

Installation for development

If you want a dev install, you should install the latest version from source with all the extra requirements for running test and generating docs.

git clone https://github.com/BlueBrain/atlas_alignment
cd atlas_alignment
pip install -e .[dev,docs]

Examples

You can find multiple examples in the documentation. Specifically, make sure to read the Building Blocks section of the docs to understand the basics.

Data

You can find example data on Zenodo. Unzip the files to ~/.atlalign/ folder so that you can use the data.py module without manual specification of paths.

Funding & Acknowledgment

This project was supported by funding to the Blue Brain Project, a research center of the Ecole polytechnique fédérale de Lausanne, from the Swiss government's ETH Board of the Swiss Federal Institutes of Technology.

COPYRIGHT (c) 2021 Blue Brain Project/EPFL