Avatar paper is an analysis project that allows researchers to see, re do, or explore figure and data of the "Patient-centric synthetic data generation, no reason to risk re-identification in biomedical data analysis". The aim of the paper is to present the avatarization method.
We really recommend to have a look to the scientific paper before to explore the repository.
Install Poetry (a Python package and dependency management tool).
This repo uses Git LFS (Large File Storage) to handle large datasets. Make sure to install it for your platform, following the instructions at https://git-lfs.github.com/
Run:
make installAll required packages will be installed using the R package librarian.
This code mainly consists in Jupyter notebooks and datasets.
It is written in Python and R.
git lfs pulldownload large file usinggit lfs, can take around 30min.- Run the following command in a terminal.
make notebook - Open notebook - such as
0.main.ipynb. - Run the cells.
datasets/
AIDS/ # original and avatarized AIDS datasets
WBCD/ # original and avatarized WBCD datasets
results_df/ # computationally expensive analysis results.
notebooks/ # analysis and graph generation
metrics/
privacy_metrics/ # multiple function used to compute avatarization metrics
figures/ # figures presented in the article
color.csv # colors for the figures
Feel free to do the analysis again or to explore avatarized datasets.
This code is licensed under the Apache-2.0 license.
- Morgan Guillaudeux [email protected]
- Olivia Rousseau [email protected]
- Julien Petot [email protected]
- Pierre-Antoine Gourraud [email protected], corresponding author.
Guillaudeux, M., Rousseau, O., Petot, J. et al. Patient-centric synthetic data generation, no reason to risk re-identification in biomedical data analysis. npj Digit. Med. 6, 37 (2023). https://doi.org/10.1038/s41746-023-00771-5