usc-mrel / DCE_direct_recon

Source codes for "Direct Estimation of Tracer-Kinetic Parameter Maps from Highly Undersampled Brain DCE-MRI", submitted to MRM

Matlab scripts for Direct Reconstruction in DCE-MRI

(c) Yi Guo, Sajan Goud Lingala, Krishna Nayak, Jul 2016.

Magnetic Resonance Engineering Laboratory

University of Southern California

Code Structure

Demo scripts

Kt_Vp_SEN_AD_3d.m Reconstruction script to demonstrate direct reconstruction from under-sampled k-space in retrospective studies.
Please download data set at https://drive.google.com/file/d/0B4nLrDuviSiWajFDV1Frc3cxR0k/view?usp=sharing for DCE50_0402.mat and https://drive.google.com/file/d/0B4nLrDuviSiWcGFPOUV1b2VrZUU/view?usp=sharing for T1_0402.mat

Functions:

conc2Ktrans.m: Convert contrast concentration to TK parameter maps.
conc2sig.m: Convert contrast concentration to signal (images).
genRGA.m: Generate randomized golden-angle radial sampling pattern.
Kt_Vp_SEN.m: Alternatively reconstruct Ktrans and Vp maps using l-bfgs.
Ktrans2conc.m: Convert Ktrans maps to contrast concentration.
sig2conc2.m: Convert signal (images) to contrast concentration.
Ktrans2sig_sen_WT.m: Cost and gradient calculation for the objective function for Ktrans.
Vp2sig_SEN_WT.m: Cost and gradient calculation for the objective funciton for Vp.
SAIF_p.m: Generate population-averaged AIF.