DCE_MOCCO

Direct model-based reconstruction of TK maps for accelerated DCE-MRI using a flexible MOCCO approach.

Demo data

please download demo phantom data from:
https://drive.google.com/file/d/0B4nLrDuviSiWT3ZKUmd0YjRwUEU/view?usp=sharing
please download demo in-vivo data from:
https://drive.google.com/file/d/0B4nLrDuviSiWXzJhLWFwN1c1ZG8/view?usp=sharing

Demo scripts

phantom_etofts_demo.m: Read pre-calculated eTofts TK maps and generated k-space (R.J Bosca et al. Phys. Med. Biol, 2016 & Y Bliesener et al. ISMRM 2017, p1909), and perform MOCCO to reconstruct TK maps from under-sampled data. Option to select different TK solver: 1. Third-party Rocketship. 2. In-house gradient solver.

AIF_TK_patlak_demo.m: Read in-vivo DCE MRI data, and retrospective under-sample the data by GOCART. Perform MOCCO to jointly reconstruct both AIF and patlak TK maps from under-sampled data.

Functions:

conc2Ktrans_Y.m: Backward modeling to convert contrast concentration to TK parameter maps.
conc2sigD.m: Convert contrast concentration to signal (image difference).
genRGA.m: Generate randomized golden-angle radial sampling pattern.
model_extended_tofts_s.m: Forward modeling from eTofts TK maps to contrast concentration.
Ktrans2conc.m: Forward modeling to convert Patlak TK maps to contrast concentration.
sig2conc2D.m: Convert signal (image difference) to contrast concentration.
multi_disp_e.m: Utility function to visualize eTofts TK parameters.
CG_recon.m: CG reconstruction of signal (image difference) from under-sampled k-space.
SAIF_p.m: Generate population-averaged AIF.