“…Many denoising methods for MRI have been proposed in the literature, these methods can be divided into three major classes [ 11 , 12 ]: (1) filtering techniques include linear filters (i.e., spatial and temporal methods) and non-linear filters (i.e., anisotropic diffusion filtering (ADF) -based methods) [ 10 ], 4th order partial differential equation (PDE) –based methods, non-local means (NLM) –based methods [ 13 ] and combination of domain and range filters (i.e., bilateral and trilateral filters); (2) transform domain methods, this class consider the curvelet and the contourlet transforms [ 14 , 15 ] and the wavelet transform based methods (i.e., wavelet thresholding, wavelet domain filter, wavelet packet analysis, adaptive multiscale product thresholding, multiwavelet and undecimated wavelet) [ 7 , 12 , 16 ]; (3) Statistical methods such as maximum likelihood estimation approach [ 17 ], Bayesian approach [ 18 ], linear minimum mean square error estimation approach, phase error estimation approach, nonparametric neighborhood statistics/estimation approach and singularity function analysis [ 11 , 18 , 19 ]. Additionally, there exist some hybrid methodologies that belong to both NLM-based methods and Statistical approaches [ 20 , 21 ].…”