The purpose of this repoistory is to provide scripts and tools to generate fitted parameters to characterize the brain ultrashort-T2* component in healthy volunteers. It is currently implemented in MATLAB and includes code for reconstructing and fitting parameter maps from brain UTE scans.
It is an open-source collaborative platform to encourage anyone to contribute
Boucneau T, Pao C, Tang S, Han M, Xu D, Henry RG, Larson PEZ. In vivo characterization of brain ultrashort-T2* components. Magnetic Resonance in Medicine. 2018 Aug;80(2):726-735. https://doi.org/10.1002/mrm.27037
- Clone the
LarsonLab/UTEMRI_Brain
github repository - Download the sample dataset which includes the raw data (pfile) and reconstructions of 18-degree, 12-degree, 6-degree scans respectively as well as a reconstructed UTE_AFI.mat file for the B1+ map correction
Run the script recon_script
which requires access ot the raw Pfiles and a list of TEs for each flip angle (from zenodo). Image Reconstruction requires the Berkeley Advanced Reconstruction Toolbox (BART) to be installed. It can be found here: https://mrirecon.github.io/bart/
This requires the Orchestra 1.7-1 sdk toolbox from GE which can be found here:https://www.gecares.com/s/GEHCForum. Extract the sdk into the same folder as the UTEMRI_Brain folder or add it to the matlab path with the command addpath(genpath'/path/to/orchestra')
Open VFA_fitting_script.m
in MATLAB and follow through the code to run and generate fitted brain uT2 parameter maps.
Approximate computation time on a 32-core CPU takes 1-1.5 hours