Reference implementation of the Reproducible Brain Charts (RBC) preprocessing protocol for anatomical, functional, derivatives, and longitudinal neuroimaging data.
This package provides a standalone implementation of the RBC preprocessing pipeline, originally developed in C-PAC. With C-PAC now deprecated, this repository serves as the maintained reference implementation using NiWrap for neuroimaging tool integration.
Reference: Shafiei et al. (2024). "Reproducible Brain Charts: An open data resource for mapping brain development and its associations with mental health." Neuron, 113(22):3758-3779.
- Anatomical preprocessing: Brain extraction (ANTs), tissue segmentation (FSL FAST), registration to MNI152
- Functional preprocessing: Motion correction, slice timing, coregistration, single-step resampling to template space
- Nuisance regression: 36-parameter and aCompCor methods with bandpass filtering
- Derivatives: ALFF/fALFF, ReHo, network centrality, atlas-based timeseries extraction
- Quality control: XCP-style QC metrics with RBC-recommended thresholds
- BIDS-compatible: Follows BIDS conventions for inputs and outputs
pip install git+https://github.com/childmindresearch/rbc.gitThe pipeline requires the following neuroimaging tools:
- AFNI: Motion correction, despiking, nuisance regression
- FSL: Tissue segmentation, registration, brain masking
- ANTs: Brain extraction, registration
from rbc.workflows import anatomical, functional
# Anatomical preprocessing
anatomical.single_session(
in_t1w="sub-01_T1w.nii.gz",
output_dir="derivatives/rbc"
)
# Functional preprocessing (in development)
functional.single_session(
in_bold="sub-01_task-rest_bold.nii.gz",
in_t1w="sub-01_T1w.nii.gz",
output_dir="derivatives/rbc"
)- API Documentation: Full API reference
Contributions are welcome! See CONTRIBUTING.md for guidelines.
# Fast tests only (for development)
pytest -m "not slow"
# All tests including integration tests
pytest
# Full pipeline tests (slow, ~30+ min)
pytest -m "full_pipeline"See tests/README.md for testing strategy and instructions.
If you use this pipeline, please cite:
@article{shafiei2024reproducible,
title={Reproducible Brain Charts: An open data resource for mapping brain development and its associations with mental health},
author={Shafiei, Golia and Esper, Natasha B and Hoffmann, Madeleine S and Ai, Lei and Chen, Andrew A and Cluce, Julia and Covitz, Sydney and Giavasis, Steven and Lane, Connor and Mehta, Kahini and Moore, Tyler M and Salo, Taylor and Tapera, Tinashe M and Calkins, Monica E and Colcombe, Stanley and Davatzikos, Christos and Gur, Raquel E and Gur, Ruben C and Pan, Pedro M and Jackowski, Andrea P and Rokem, Ariel and Rohde, Luis A and Shinohara, Russell T and Tottenham, Nim and Zuo, Xi-Nian and Cieslak, Matthew and Franco, Alexandre R and Kiar, Gregory and Salum, Giovanni A and Milham, Michael P and Satterthwaite, Theodore D},
journal={Neuron},
volume={113},
number={22},
pages={3758--3779},
year={2024},
publisher={Elsevier},
doi={10.1016/j.neuron.2024.08.026}
}Reproducible Brain Charts (RBC) is an open resource integrating data from 5 large studies of brain development in youth from three continents (N = 6,346). The resource provides:
- Harmonized psychiatric phenotypes using bifactor models
- Quality-assured neuroimaging data processed with consistent pipelines
- All data openly shared via the International Neuroimaging Data-sharing Initiative (INDI)
RBC facilitates large-scale, reproducible, and generalizable research in developmental and psychiatric neuroscience.
This implementation is based on the RBC protocol described in Shafiei et al. (2024) and originally implemented in C-PAC. Development is supported by the Child Mind Institute.