IRUS Total

DAGAN: deep de-aliasing generative adversarial networks for fast compressed sensing MRI reconstruction

File Description SizeFormat 
08233175.pdfPublished version3.75 MBAdobe PDFView/Open
Title: DAGAN: deep de-aliasing generative adversarial networks for fast compressed sensing MRI reconstruction
Authors: Yang, G
Yu, S
Hao, D
Slabaugh, G
Dragotti, PL
Ye, X
Liu, F
Arridge, S
Keegan, J
Guo, Y
Firmin, D
Item Type: Journal Article
Abstract: Compressed Sensing Magnetic Resonance Imaging (CS-MRI) enables fast acquisition, which is highly desirable for numerous clinical applications. This can not only reduce the scanning cost and ease patient burden, but also potentially reduce motion artefacts and the effect of contrast washout, thus yielding better image quality. Different from parallel imaging based fast MRI, which utilises multiple coils to simultaneously receive MR signals, CS-MRI breaks the Nyquist-Shannon sampling barrier to reconstruct MRI images with much less required raw data. This paper provides a deep learning based strategy for reconstruction of CS-MRI, and bridges a substantial gap between conventional non-learning methods working only on data from a single image, and prior knowledge from large training datasets. In particular, a novel conditional Generative Adversarial Networks-based model (DAGAN) is proposed to reconstruct CS-MRI. In our DAGAN architecture, we have designed a refinement learning method to stabilise our U-Net based generator, which provides an endto-end network to reduce aliasing artefacts. To better preserve texture and edges in the reconstruction, we have coupled the adversarial loss with an innovative content loss. In addition, we incorporate frequency domain information to enforce similarity in both the image and frequency domains. We have performed comprehensive comparison studies with both conventional CSMRI reconstruction methods and newly investigated deep learning approaches. Compared to these methods, our DAGAN method provides superior reconstruction with preserved perceptual image details. Furthermore, each image is reconstructed in about 5 ms, which is suitable for real-time processing.
Issue Date: 1-Jun-2018
Date of Acceptance: 18-Dec-2017
URI: http://hdl.handle.net/10044/1/55724
DOI: 10.1109/TMI.2017.2785879
ISSN: 0278-0062
Publisher: Institute of Electrical and Electronics Engineers
Start Page: 1310
End Page: 1321
Journal / Book Title: IEEE Transactions on Medical Imaging
Volume: 37
Issue: 6
Copyright Statement: © 2017 IEEE. This work is licensed under a Creative Commons Attribution 3.0 License. For more information, see h ttp://creativecommons.org/licenses/by/3.0/
Sponsor/Funder: British Heart Foundation
Funder's Grant Number: PG/16/78/32402
Keywords: Science & Technology
Life Sciences & Biomedicine
Computer Science, Interdisciplinary Applications
Engineering, Biomedical
Engineering, Electrical & Electronic
Imaging Science & Photographic Technology
Radiology, Nuclear Medicine & Medical Imaging
Computer Science
Compressed sensing
magnetic resonance imaging (MRI)
fast MRI
deep learning
generative adversarial networks (GAN)
inverse problems
Data Compression
Deep Learning
Magnetic Resonance Imaging
Magnetic Resonance Imaging
Data Compression
Deep Learning
Nuclear Medicine & Medical Imaging
08 Information and Computing Sciences
09 Engineering
Publication Status: Published
Online Publication Date: 2017-12-21
Appears in Collections:Electrical and Electronic Engineering
National Heart and Lung Institute
Faculty of Medicine
Faculty of Engineering