Skip to content

Latest commit

 

History

History
50 lines (35 loc) · 1.44 KB

README.md

File metadata and controls

50 lines (35 loc) · 1.44 KB

Analysis

Binder

Registering images

Prerequesites

  • ANTs (tested with version 2.3.3.dev168-g29bdf)
  • Python 3.7
  • Clone this repository and install its requirements:
    git clone https://github.com/rrsg2020/analysis.git rrsg2020/analysis
    cd rrsg2020/analysis
    pip install -r requirements.txt

Download datasets and pool into one folder

Run the commands:

python make_pooled_datasets.py configs/3T_NIST_T1maps.json 3T_NIST_T1maps
python make_pooled_datasets.py configs/3T_NIST.json 3T_NIST

Note: Labels and a phantom mask are already included in the T1 map OSF dataset

Register datasets to the reference

Run the registration script:

python register_t1maps_nist.py -j configs/3T_NIST.json -p 3T_NIST_pooled/ 3T_NIST_T1maps_pooled/

Note: the registration script will download the reference mask (e.g. for the NIST phantom) and will create labels for the initial affine transformation.

Installation instructions for ARM processors (e.g. Apple M1 chip on MacBook Pros 2020 and later).

conda env create -f environment_ARM.yml
conda activate analysis_arm 
python setup_ARM.py install
python -m ipykernel install --user --name=analysis_arm
pytest