Literature DB >> 33107127

Deep neural network for water/fat separation: Supervised training, unsupervised training, and no training.

Ramin Jafari1,2, Pascal Spincemaille2, Jinwei Zhang1,2, Thanh D Nguyen2, Xianfu Luo2,3, Junghun Cho1,2, Daniel Margolis2, Martin R Prince2, Yi Wang1,2.   

Abstract

PURPOSE: To use a deep neural network (DNN) for solving the optimization problem of water/fat separation and to compare supervised and unsupervised training.
METHODS: The current T 2 ∗ -IDEAL algorithm for solving water/fat separation is dependent on initialization. Recently, DNN has been proposed to solve water/fat separation without the need for suitable initialization. However, this approach requires supervised training of DNN using the reference water/fat separation images. Here we propose 2 novel DNN water/fat separation methods: 1) unsupervised training of DNN (UTD) using the physical forward problem as the cost function during training, and 2) no training of DNN using physical cost and backpropagation to directly reconstruct a single dataset. The supervised training of DNN, unsupervised training of DNN, and no training of DNN methods were compared with the reference T 2 ∗ -IDEAL.
RESULTS: All DNN methods generated consistent water/fat separation results that agreed well with T 2 ∗ -IDEAL under proper initialization.
CONCLUSION: The water/fat separation problem can be solved using unsupervised deep neural networks.
© 2020 International Society for Magnetic Resonance in Medicine.

Entities:  

Keywords:  deep learning; label free; unsupervised; water/fat separation

Year:  2020        PMID: 33107127      PMCID: PMC7809709          DOI: 10.1002/mrm.28546

Source DB:  PubMed          Journal:  Magn Reson Med        ISSN: 0740-3194            Impact factor:   4.668


  18 in total

1.  Multicoil Dixon chemical species separation with an iterative least-squares estimation method.

Authors:  Scott B Reeder; Zhifei Wen; Huanzhou Yu; Angel R Pineda; Garry E Gold; Michael Markl; Norbert J Pelc
Journal:  Magn Reson Med       Date:  2004-01       Impact factor: 4.668

2.  Proton density fat-fraction: a standardized MR-based biomarker of tissue fat concentration.

Authors:  Scott B Reeder; Houchun H Hu; Claude B Sirlin
Journal:  J Magn Reson Imaging       Date:  2012-07-06       Impact factor: 4.813

3.  Fidelity imposed network edit (FINE) for solving ill-posed image reconstruction.

Authors:  Jinwei Zhang; Zhe Liu; Shun Zhang; Hang Zhang; Pascal Spincemaille; Thanh D Nguyen; Mert R Sabuncu; Yi Wang
Journal:  Neuroimage       Date:  2020-01-22       Impact factor: 6.556

4.  Simultaneous phase unwrapping and removal of chemical shift (SPURS) using graph cuts: application in quantitative susceptibility mapping.

Authors:  Jianwu Dong; Tian Liu; Feng Chen; Dong Zhou; Alexey Dimov; Ashish Raj; Qiang Cheng; Pascal Spincemaille; Yi Wang
Journal:  IEEE Trans Med Imaging       Date:  2014-10-08       Impact factor: 10.048

5.  Quantitative susceptibility mapping of the spine using in-phase echoes to initialize inhomogeneous field and R2* for the nonconvex optimization problem of fat-water separation.

Authors:  Yihao Guo; Zhe Liu; Yan Wen; Pascal Spincemaille; Honglei Zhang; Ramin Jafari; Shun Zhang; Sarah Eskreis-Winkler; Kelly M Gillen; Peiwei Yi; Qianjin Feng; Yanqiu Feng; Yi Wang
Journal:  NMR Biomed       Date:  2019-08-19       Impact factor: 4.044

Review 6.  Linearity, Bias, and Precision of Hepatic Proton Density Fat Fraction Measurements by Using MR Imaging: A Meta-Analysis.

Authors:  Takeshi Yokoo; Suraj D Serai; Ali Pirasteh; Mustafa R Bashir; Gavin Hamilton; Diego Hernando; Houchun H Hu; Holger Hetterich; Jens-Peter Kühn; Guido M Kukuk; Rohit Loomba; Michael S Middleton; Nancy A Obuchowski; Ji Soo Song; An Tang; Xinhuai Wu; Scott B Reeder; Claude B Sirlin
Journal:  Radiology       Date:  2017-09-11       Impact factor: 11.105

7.  Improving chemical shift encoded water-fat separation using object-based information of the magnetic field inhomogeneity.

Authors:  Samir D Sharma; Nathan S Artz; Diego Hernando; Debra E Horng; Scott B Reeder
Journal:  Magn Reson Med       Date:  2014-02-28       Impact factor: 4.668

8.  Water-fat separation and parameter mapping in cardiac MRI via deep learning with a convolutional neural network.

Authors:  James W Goldfarb; Jason Craft; J Jane Cao
Journal:  J Magn Reson Imaging       Date:  2019-01-30       Impact factor: 4.813

9.  Quantitative susceptibility mapping in the abdomen as an imaging biomarker of hepatic iron overload.

Authors:  Samir D Sharma; Diego Hernando; Debra E Horng; Scott B Reeder
Journal:  Magn Reson Med       Date:  2014-09-08       Impact factor: 4.668

10.  Separation of water and fat signal in whole-body gradient echo scans using convolutional neural networks.

Authors:  Jonathan Andersson; Håkan Ahlström; Joel Kullberg
Journal:  Magn Reson Med       Date:  2019-04-29       Impact factor: 4.668

View more
  2 in total

1.  QQ-NET - using deep learning to solve quantitative susceptibility mapping and quantitative blood oxygen level dependent magnitude (QSM+qBOLD or QQ) based oxygen extraction fraction (OEF) mapping.

Authors:  Junghun Cho; Jinwei Zhang; Pascal Spincemaille; Hang Zhang; Simon Hubertus; Yan Wen; Ramin Jafari; Shun Zhang; Thanh D Nguyen; Alexey V Dimov; Ajay Gupta; Yi Wang
Journal:  Magn Reson Med       Date:  2021-10-31       Impact factor: 3.737

2.  Computed Tomography Image Features under Deep Learning Algorithm Applied in Staging Diagnosis of Bladder Cancer and Detection on Ceramide Glycosylation.

Authors:  Yisheng Xu; Jianghua Lou; Zhiqin Gao; Ming Zhan
Journal:  Comput Math Methods Med       Date:  2022-01-07       Impact factor: 2.238

  2 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.