Abstract

Info

Author(s):: CHAI Qinghuan; SU Guanqun; NIE Shengdong; Institute of Medical Imaging Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China

Keywords:: ; magnetic resonance imaging; partial K-space; compressed sensing; reconstruction algorithm; review

Abstract:: Magnetic resonance imaging (MRI) is one of the necessary imaging methods for acquiring clinical images, but its inherent slow data acquisition capability causes a long time for imaging. At present, many high-efficient imaging algorithms are proposed to shorten the time for MRI, such as semi-Fourier imaging and MRI based on compressed sensing. Semi-Fourier imaging, one of the effective partial K-space reconstruction technologies, adopts only more than half of K-space data for image reconstruction, which not only improves the imaging speed, but also reduces the motion artifacts. MRI based on compressed sensing theory can reconstruct images using only 25% to 30% of K-space data. Compared with other imaging technologies, compressed sensing-based MRI can obtain MRI images of higher quality in the same scan time and accelerate imaging in the same spatial resolution. Herein we review the principles of several semi-Fourier imaging algorithms and introduce the principles of compressed sensing theory combined with MRI, including sparse representation of MRI image, sampling trajectory design of K-space, selection of reconstruction algorithm and so on.