APART-QSM is a new proposed QSM separation method using an iterative data fitting scheme.
APART-QSM can separate paramagnetic and diamagnetic susceptibility components on in vivo and ex vivo data.
APART-QSM can handle an arbitrary number of complex GRE data input measurements to provide high-quality QSM separation maps with more faithful tissue delineation of the small brain sub-regions.
This repository contains two implementations of APART-QSM method based on single-orientation and multiple orientation data.
Li, Z., Feng, R., Liu, Q., Feng, J., Lao, G., Zhang, M., ... & Wei, H. (2023). APART-QSM: An improved sub-voxel quantitative susceptibility mapping for susceptibility source separation using an iterative data fitting method. Neuroimage, 274, 120148.
preprocessing on phase data and QSM initialization
Nifti toolbox
saving reconstruction results
brain mask generation and registration for multi-orientation data
Recommended MATLAB version: R2019a, R2020a
APART_QSM_single_ori_demo.m is a demo for susceptibility separation using single orientation data.
APART_QSM_multi_ori_demo.m is a demo for susceptibility separation using multiple orientation data.
For data acquired with 3D multi-echo GRE sequences, you should do the following preprocessing steps before running STI reconstruction codes in this repository:
(1) Extract the tissue mask from magnitude images using FSL Bet.
(2) Unwrap the raw phase data using Laplacian-based phase unwrapping in STI_Suite V3.0 toolbox.
(3) Divide the background and local phase using VSHARP in STI_Suite V3.0 toolbox.
(4) Co-register the magnitude images at different orientations to a reference orientation (supine position) using FSL FLIRT.
(5) Apply the transform matrix to the corresponding local phase and calculate magnetic field direction (obtain B0_dirs).
(6) Once the above steps are completed, you can run demos successfully as long as Nifti toolbox is installed and added to the path.
Our software license is exclusively offered for academic research purposes and is expressly NOT intended for commercial or clinical use. To acquire the license, please email sjtu.amri@gmail.com, providing your full name, affiliated institution, and the reason for renewal.