Dual-domain convolutional neural networks for improving structural information in 3 T MRI.

Journal: Magnetic resonance imaging
Published Date: Jun 5, 2019

Abstract

We propose a novel dual-domain convolutional neural network framework to improve structural information of routine 3 T images. We introduce a parameter-efficient butterfly network that involves two complementary domains: a spatial domain and a frequency domain. The butterfly network allows the interaction of these two domains in learning the complex mapping from 3 T to 7 T images. We verified the efficacy of the dual-domain strategy and butterfly network using 3 T and 7 T image pairs. Experimental results demonstrate that the proposed framework generates synthetic 7 T-like images and achieves performance superior to state-of-the-art methods.

Authors

  • Yongqin Zhang
    School of Information Science and Technology, Northwest University, Xi'an 710127, China; Department of Radiology and BRIC, University of North Carolina at Chapel Hill, NC 27599, USA. Electronic address: zhangyongqin@nwu.edu.cn.
  • Pew-Thian Yap
    Department of Radiology and BRIC, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA.
  • Liangqiong Qu
  • Jie-Zhi Cheng
    National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, School of Medicine, Shenzhen University, Shenzhen, Guangdong 518060, P.R. China.
  • Dinggang Shen
    School of Biomedical Engineering, ShanghaiTech University, Shanghai, China.

Keywords

External Resources

Popular Topics
Recent Journals