cs-3dcine-recon

MATLAB-based code for reconstructing 3d cine whole-heart data acquired using the AUMC PROUD patch

Description:

Compressed sensing (CS) based reconstruction of raw Philips data is performed using MATLAB-based implementations of ReconFrame and bart toolboxes. This code is specifically designed for reconstructing data acquired using the PROspective Undersampling in multiple Dimensions (PROUD) patch built and validated at Amsterdam UMC, location AMC.

Required MATLAB toolboxes:

• Image Processing Toolbox
• Parallel Computing Toolbox

Additional toolboxes:

• bart, either on a linux system or using a Windows-based install.
• MRecon from Gyrotools.

Getting started:

cs-3dcine-recon requires the definition of a reconstruction protocol file (.pro.dat). This file contains all information needed from the reconstruction code. To begin, open MATLAB at the cs-3dcine-recon-master folder and type:

run_defineprofile.m