BIDScramble is a versatile tool that generates scrambled or pseudo-random BIDS datasets from existing BIDS-structured data. It carefully preserves the statistical distributions of user-specified variables while maintaining the effects of interest, ensuring that the synthetic data remains analytically useful. Crucially, the output contains no or minimal indirect personal data, making it privacy-compliant and untraceable to original subjects.
This tool enables researchers to safely explore, prototype, and test their analysis pipelines on realistic yet anonymized data. Since the scrambled datasets retain key statistical properties, pipelines should behave similarly to how they would on real data — accelerating development without compromising privacy.
- Explore the full documentation on Read the Docs
- Access the open-source code on GitHub (licensed under GPL-3.0-or-later)
- BIDScramble is distributed without any warranty. See the copyright notice for more details.
BIDScramble is being developed at the Donders Institute of the Radboud University as part of the EOSC SIESTA project. Funding was received from the European Union’s Horizon Research and Innovation program under Grant agreement No. 101131957.
