Brain-TokenGT

This is the official repository for "Beyond the Snapshot: Brain Tokenized Graph Transformer for Longitudinal Brain Functional Connectome Embedding" (MICCAI 2023) model illustration figure

Dependencies

The framework needs the following dependencies:

numpy==1.24.2
optuna==3.1.0
PyYAML==6.0
scikit_learn==1.2.2
scipy==1.9.1
torch==2.0.0
torch_geometric==2.2.0
tqdm==4.64.1

Datasets

• ADNI: https://adni.loni.usc.edu/
• OASIS: https://www.oasis-brains.org/

We used brain FC metrics derived from ADNI and OASIS-3 resting state fMRI datasets, with preprocessing pipelines following:

Kong, R., Li, J., Orban, C., Sabuncu, M.R., Liu, H., Schaefer, A., Sun, N., Zuo, X.N., Holmes, A.J., Eickhoff, S.B., et al.: Spatial topography of individual-specific cortical networks predicts human cognition, personality, and emotion. Cerebral cortex 29(6), 2533–2551 (2019)

Li, J., Kong, R., Li´egeois, R., Orban, C., Tan, Y., Sun, N., Holmes, A.J., Sabuncu, M.R., Ge, T., Yeo, B.T.: Global signal regression strengthens association between resting-state functional connectivity and behavior. NeuroImage 196, 126– 141 (2019)

Installation

1. Clone the repository: git clone https://github.com/ZijianD/Brain-TokenGT.git
2. Change to the project directory: cd Brain-TokenGT
3. Install the dependencies: pip install -r requirements.txt

Usage

python main_optuna.py # you may modify config.py to change the hyperparameter setup

References

Our implementation uses code from the following repositories:

• EvolveGCN
• tokengt