MRI reconstruction via non-convex total variation regularization

The demos implement the experiments given in the paper "Magnetic Resonance Imaging reconstruction via non-convex total variation regularization (MCTV_L2)".

After unzipping the supplementary file, please add all sub-directories into your MATLAB paths.

• 'main.m' give a simple example to illustrate the efficiency of MCTV_L2.

• 'rec_mctv2.m' and 'shrink_mctv2_1.m' are the main part of MCTV_L2.

• 'rec_mctv.m' and 'shrink_mctv.m' are from paper "Convex MR brain image reconstruction via non-convex total variation minimization" by Liu et.al..

• 'mask_radial.m' is used to generate radial sampling templates. Template size and number of trajectory lines can be chosen.

• 'mask_random.m' is used to generate random sampling templates. genPDF.m and gengenSampling.m are subroutines of mask_random.m (© 2007 Michael Lustig). Template size, sampling radius and sampling rate can be chosen.

• 'mask_Cartesian.m' is used to generate Cartesian templates. Template size and number of readout lines can be chosen.

Constacts: [email protected]