We use NiChart_DLMUSE for skull-stripping (s3 masked folder) and flirt for registration, though you can use any skull-stripping model and achieve comparable results.
You can install NiChart_DLMUSE from source:
git clone https://github.com/CBICA/NiChart_DLMUSE
cd NiChart_DLMUSE
pip install -e .To install flirt, please read flirt's documentation.
Tip
We suggest to use a different python virtualenv for preprocessing and Brain2Vec inference to avoid dependency issues with DLMUSE.
To run the pre-processing pipeline, just do:
python brain2vec_preprocess.py --in_dir ... --out_dir ... --device cpu/cudaIn out_dir/brain2vec_aligned_scans, you will find the Brain2Vec pre-processed scans ready for inference.