Meeting Banner
Abstract #2270

Improving Edge Recovery in Udersampled MRI Reconstruction

Zhong Chen1, Changwei Hu1, Xiaobo Qu1, Lijun Bao1, Shuhui Cai1

1Department of Electronic Science, Fujian Key Laboratory of Plasma and Magnetic Resonance, Xiamen University, Xiamen, Fujian, China

Undersampling k-space is an effective way to reduce acquisition time in MRI. However, this will introduce significant aliasing artifacts, and blur edges in the reconstructed magnetic resonance image. The edges usually contain important information for the clinical diagnosis. In this work, we propose a method to recover edges from undersampled MRI by incoporating a weighting matrix to the l1 norm sparsity regularization term. Compared with conventional compressive sensing methods, the proposed method yields better edge recovery, and requires fewer k-space measurements to achieve acceptable reconstruction quality.

Abstract PDF

Keywords

acceptable achieve acknowledgment addition adopted aliasing among applications artifacts beck better blur caliber china clearer clinical coefficient coefficients combining component components compressed considering consistency contain continuity contribution correlation curve curves decomposition define defined denotes designed diagnosis directional divided domain edge edges electronic employed error evaluate fall fast features fewer filters focal formulation fully function generated give good governing grows highly important improving incorporated indicate indicates intensity introduced iterative labeled laboratory like local lowest margin matrix maximal measure measured minimization morphological much nearly norm normalized observe orientation passing patch patches pattern people performance plasma produced proposed quality quantify reconstructed reconstruction reconstructions recovery reflects regularization relatively representation represents republic requires respectively rice sampled sampling scanner scheme science scoring sensing shaped sharing simulations soft space sparse sparsity spatial spline square strength supported suppose thank think thresholding toolbox transform transition trial tumor useful usually vectorizing vectors versus vessel wavelet window windows yields zero zoom