Craniosource-GAN-PCA-SSM

Overview

This is a repository to use a GAN, PCA, and statistical shape models to create craniosynostosis distance maps, for example for the classification of craniosynostosis on purely synthetic data.

Requirements and installation

We used Python version 3.7, but any reasonably recent version of Python3 should be fine. Create a virtual environment to install the required packages using the requirements file:

# Create the virtual environment
mkdir -p $HOME/.venv
python3 -m venv $HOME/.venv/map_creator

# Activate the environment
source $HOME/.venv/map_creator/bin/activate

# Install the dependencies
pip install -r pip_requirements.txt

Alternatively, the dependencies are listed below and installing them via pip should suffice to re-create the virtual environment.

• numpy
• scikit-learn
• scikit-image
• vtk
• torch

Run demo

The python code is cross-platform and should run everywhere. The demo is a simple shell script that just calles the script with different input parametern and should work anywhere where bash can be installed. On Windows, we recommend WSL.

bash ./demo.sh

Alternatively, just take a look at the demo file to figure out how to call the scripts.

License

All source code is subject to the terms of the General Public License 3.0.