MRI reconstruction with enhanced self-similarity using graph convolutional network.

Journal: BMC medical imaging
Published Date:

Abstract

BACKGROUND: Recent Convolutional Neural Networks (CNNs) perform low-error reconstruction in fast Magnetic Resonance Imaging (MRI). Most of them convolve the image with kernels and successfully explore the local information. Nonetheless, the non-local image information, which is embedded among image patches relatively far from each other, may be lost due to the limitation of the receptive field of the convolution kernel. We aim to incorporate a graph to represent non-local information and improve the reconstructed images by using the Graph Convolutional Enhanced Self-Similarity (GCESS) network.

Authors

  • Qiaoyu Ma
    School of Ocean Information Engineering, Jimei University, Xiamen, China.
  • Zongying Lai
    School of Information Engineering, Jimei University, Xiamen, China.
  • Zi Wang
    Clinical Medical College, Yangzhou University, 225009 Yangzhou, Jiangsu, China.
  • Yiran Qiu
    School of Ocean Information Engineering, Jimei University, Xiamen, China.
  • Haotian Zhang
  • Xiaobo Qu