Research of MRI Reconstruction Method by Using De-aliasing Wasserstein Generative Adversarial Networks with Gradient Penalty

被引:0
作者
Yuan Z.-H. [1 ]
Jiang M.-F. [1 ]
Li Y. [1 ]
Zhi M.-H. [1 ]
Zhu Z.-J. [2 ]
机构
[1] School of Information Science and Technology, Zhejiang Sci-Tech University, Hangzhou
[2] Department of Cardiology, The 117th Hospital of PLA, Hangzhou
来源
Tien Tzu Hsueh Pao/Acta Electronica Sinica | 2020年 / 48卷 / 10期
关键词
Image reconstruction; Magnetic resonance imaging (MRI); Perceptual loss; WGAN-GP;
D O I
10.3969/j.issn.0372-2112.2020.10.002
中图分类号
学科分类号
摘要
In this paper,we propose an improved Wasserstein generative adversarial network (WGAN),de-aliasing Wasserstein generative adversarial network with Gradient Penalty (DAWGAN-GP),for magnetic resonance imaging (MRI) reconstruction.This method uses WGAN to replace the traditional GAN,and combined the gradient penalty to improve the training speed and to solve the slow convergence problem of WGAN.In addition,for better preservation of the fine structures in the reconstructed images,we incorporate perceptual loss,pixel loss and frequency loss into the loss function for training the network.Compared with other state-of-the-art deep learning methods for MR images reconstruction,DAWGAN-GP method outperforms all other methods and can provide superior reconstruction with improved peak signal to noise ratio (PSNR) and better structural similarity index measure (SSIM). © 2020, Chinese Institute of Electronics. All right reserved.
引用
收藏
页码:1883 / 1890
页数:7
相关论文
共 21 条
  • [1] Donoho D L., Compressed sensing[J], IEEE Transactions on Information Theory, 52, 4, pp. 1289-1306, (2006)
  • [2] Lustig M, Donoho D L, Santos J M, Et al., Compressed sensing MRI[J], IEEE Signal Processing Magazine, 25, 2, pp. 72-82, (2008)
  • [3] Yang Y, Liu F, Li M, Et al., Pseudo-polar Fourier transform based compressed sensing MRI[J], IEEE Transactions on Biomedical Engineering, 64, 4, pp. 816-825, (2017)
  • [4] JIANG Ming-feng, LU Liang, WU Long, Et al., The research of MRI reconstruction method by using weighted Schatten P-norm minimization [J], Acta Electronica Sinica, 47, 4, pp. 784-790, (2019)
  • [5] Liu S, Cao J, Wu G, Et al., CS-MRI reconstruction via group-based Eigenvalue decomposition and estimation, Neurocomputing, 283, pp. 166-180, (2018)
  • [6] Feng L, Srichai M B, Lim R P, Et al., Highly accelerated real-time cardiac cine MRI using k-t sparse-sense[J], Magnetic Resonance in Medicine, 70, 1, pp. 64-74, (2013)
  • [7] Feng L, Grimm R, Block K T, Et al., Golden-angle radial sparse parallel MRI:Combination of compressed sensing,parallel imaging,and golden-angle radial sampling for fast and flexible dynamic volumetric MRI[J], Magnetic Resonance in Medicine, 72, 3, pp. 707-717, (2015)
  • [8] Wang S, Tan S, Gao Y, Et al., Learning joint-sparse codes for calibration-free parallel MR imaging (LINDBERG)[J], IEEE Transactions on Medical Imaging, 37, 1, pp. 251-261, (2018)
  • [9] Huang Y, Paisley J, Lin Q, Et al., Bayesian nonparametric dictionary learning for compressed sensing MRI[J], IEEE Transactions on Image Processing, 23, 12, pp. 5007-5019, (2013)
  • [10] Zhan Z, Cai J F, Guo D, Et al., Fast multi-class dictionaries learning with geometrical directions in MRI reconstruction[J], IEEE Transactions on Biomedical Engineering, 63, 9, pp. 1850-1861, (2016)