/LARO-QSM

Primary LanguagePython

LARO: Learned Acquisition and Reconstruction Optimization to accelerate Quantitative Susceptibility Mapping

Code of LARO for multi-echo gradient echo (mGRE) acceleration for Quantitative Susceptibility Mapping (NeuroImage 2023, MICCAI 2021 and MLMIR@MICCAI 2020)

Network architecture of LARO. (a) deep ADMM reconstruction. (b) sampling pattern optimization module. (c) temporal feature fusion module: Network architecture of LARO. (a): deep ADMM reconstruction. (b): sampling pattern optimization module. (c): temporal feature fusion module.

Illustration of (a) the proposed segmented k-space ordering strategy of ten echoes and (b) pulse sequence design: Illustration of (a): the proposed segmented k-space ordering strategy of ten echoes and (b): pulse sequence design.

TFF reconstructions on prospectively under-sampled raw k-space data of one healthy subject with acceleration factor R=8: TFF reconstructions on prospectively under-sampled raw k-space data of one healthy subject with acceleration factor R=8.

To simultaneously update the multi-echo sampling pattern and image reconstruction network, run:

python main_multi_echo_MS.py --flag_train=1 --K=2 --loupe=2

Please contact jz853@cornell.edu for questions and data sharing.