Compressed sensing longitudinal MRI based on wavelet and shearlet
KANG Ruirui
CAO Bin
WANG Bin
YAN Long
QU Gangrong
Abstract：In order to overcome the deficiencies that the traditional two-dimensional wavelet trans-form can not provide the optimal representation for MR images and to reduce the scanning time of MR images,a new regularization model and the corresponding reconstruction algorithm(frsLCS-MRI)are proposed based on the joint sparse transform and similarity prior.In order to verify the effectiveness of the joint sparse transform and the proposed algorithm frsLCS-MRI,we compare it with the LACS-MRI algorithm based on single sparse transform and similarity by using two medical datasets.Numerical results indicate that the proposed method frsLCS-MRI can recon-struct MR images with higher accuracy and higher signal to noise ratio.Compared with the meth-od LACS-MRI using only wavelet transform,or shearlet transform,more detail information such as boundaries,corners,contours,can be reconstructed by the proposed method frsLCS-MRI using the joint sparse transform.Moreover,exploiting similarity based on reference image saves acquisition time.
Keywords：image processingmagnetic resonance imaging(MRI)compressed sensingsimilaritywaveletshearlet
Publication Date：2017-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：6( 127-132 )