Skip to main content
Advertisement

Main menu

  • Home
  • Content
    • Current Issue
    • Publication Preview--Ahead of Print
    • Past Issue Archive
    • Case of the Week Archive
    • Classic Case Archive
    • Case of the Month Archive
  • For Authors
  • About Us
    • About AJNR
    • Editors
    • American Society of Neuroradiology
  • Submit a Manuscript
  • Podcasts
    • Subscribe on iTunes
    • Subscribe on Stitcher
  • More
    • Subscribers
    • Permissions
    • Advertisers
    • Alerts
    • Feedback
  • Other Publications
    • ajnr

User menu

  • Subscribe
  • Alerts
  • Log in

Search

  • Advanced search
American Journal of Neuroradiology
American Journal of Neuroradiology

American Journal of Neuroradiology

  • Subscribe
  • Alerts
  • Log in

Advanced Search

  • Home
  • Content
    • Current Issue
    • Publication Preview--Ahead of Print
    • Past Issue Archive
    • Case of the Week Archive
    • Classic Case Archive
    • Case of the Month Archive
  • For Authors
  • About Us
    • About AJNR
    • Editors
    • American Society of Neuroradiology
  • Submit a Manuscript
  • Podcasts
    • Subscribe on iTunes
    • Subscribe on Stitcher
  • More
    • Subscribers
    • Permissions
    • Advertisers
    • Alerts
    • Feedback
  • Follow AJNR on Twitter
  • Visit AJNR on Facebook
  • Follow AJNR on Instagram
  • Join AJNR on LinkedIn
  • RSS Feeds
Research ArticleAdult Brain
Open Access

Improving the Quality of Synthetic FLAIR Images with Deep Learning Using a Conditional Generative Adversarial Network for Pixel-by-Pixel Image Translation

A. Hagiwara, Y. Otsuka, M. Hori, Y. Tachibana, K. Yokoyama, S. Fujita, C. Andica, K. Kamagata, R. Irie, S. Koshino, T. Maekawa, L. Chougar, A. Wada, M.Y. Takemura, N. Hattori and S. Aoki
American Journal of Neuroradiology February 2019, 40 (2) 224-230; DOI: https://doi.org/10.3174/ajnr.A5927
A. Hagiwara
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
cDepartment of Radiology (A.H., R.I., S.K., T.M.), Graduate School of Medicine, University of Tokyo, Tokyo, Japan
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for A. Hagiwara
Y. Otsuka
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
dMilliman Inc (Y.O.). Tokyo, Japan
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for Y. Otsuka
M. Hori
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for M. Hori
Y. Tachibana
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
eApplied MRI Research (Y.T.), Department of Molecular Imaging and Theranostics, National Institute of Radiological Sciences, Chiba, Japan
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for Y. Tachibana
K. Yokoyama
bNeurology (K.Y., N.H.), Juntendo University School of Medicine, Tokyo, Japan
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for K. Yokoyama
S. Fujita
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for S. Fujita
C. Andica
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for C. Andica
K. Kamagata
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for K. Kamagata
R. Irie
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
cDepartment of Radiology (A.H., R.I., S.K., T.M.), Graduate School of Medicine, University of Tokyo, Tokyo, Japan
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for R. Irie
S. Koshino
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
cDepartment of Radiology (A.H., R.I., S.K., T.M.), Graduate School of Medicine, University of Tokyo, Tokyo, Japan
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for S. Koshino
T. Maekawa
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
cDepartment of Radiology (A.H., R.I., S.K., T.M.), Graduate School of Medicine, University of Tokyo, Tokyo, Japan
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for T. Maekawa
L. Chougar
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
fDepartment of Radiology (L.C.), Hopital Saint-Joseph, Paris, France; and Department of Radiological Sciences.
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for L. Chougar
A. Wada
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for A. Wada
M.Y. Takemura
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for M.Y. Takemura
N. Hattori
bNeurology (K.Y., N.H.), Juntendo University School of Medicine, Tokyo, Japan
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for N. Hattori
S. Aoki
aFrom the Departments of Radiology (A.H., Y.O., M.H., Y.T., S.F., C.A., K.K., R.I., S.K., T.M., L.C., A.W., M.Y.T., S.A.)
  • Find this author on Google Scholar
  • Find this author on PubMed
  • Search for this author on this site
  • ORCID record for S. Aoki
  • Article
  • Figures & Data
  • Info & Metrics
  • References
  • PDF
Loading

Article Figures & Data

Figures

  • Tables
  • Fig 1.
    • Download figure
    • Open in new tab
    • Download powerpoint
    Fig 1.

    A, Illustration describing the generator. B, Illustration describing the discriminator. C, The framework describing the training phase of our proposed cGAN model for improving the synthetic FLAIR image quality. BN indicates batch normalization; Conv, convolution; eLU, exponential linear unit; ch, channel.

  • Fig 2.
    • Download figure
    • Open in new tab
    • Download powerpoint
    Fig 2.

    Synthetic FLAIR (A), DL-FLAIR (B), and conventional FLAIR (C) images of a representative patient. The overall image contrast of the DL-FLAIR image is more similar to that of the conventional FLAIR image than the contrast of the synthetic FLAIR image, while preserving the lesion contrast. The NRMSE maps of synthetic FLAIR (D) and DL-FLAIR (E) images against conventional FLAIR images are also shown. The NRMSE in the intracranial tissues is much larger in the synthetic FLAIR image than in the DL-FLAIR image. Note that the parenchymal surface shows patchy high NRMSE values on the synthetic FLAIR image, which are reduced but still visible in the DL-FLAIR image.

  • Fig 3.
    • Download figure
    • Open in new tab
    • Download powerpoint
    Fig 3.

    A, Granular artifacts in the CSF on a synthetic FLAIR image. B, The artifact is almost invisible (successfully deleted) on the DL-FLAIR image. C, Conventional FLAIR image is shown for comparison.

  • Fig 4.
    • Download figure
    • Open in new tab
    • Download powerpoint
    Fig 4.

    Magnified images from Fig 2. Synthetic FLAIR (A), DL-FLAIR (B), and conventional FLAIR (C) images are shown. Sulci are wider in B and C in some areas than they are in A (white arrows). However, for areas with tight sulci on the conventional FLAIR image (C), the sulci are tighter and more hyperintense on both synthetic FLAIR (A) and DL-FLAIR (B) images than they are on the conventional FLAIR image (black arrows).

Tables

  • Figures
  • The PSNR and NRMSE of synthetic FLAIR and DL-FLAIR images against conventional FLAIR images in various regionsa

    Synthetic FLAIR ImagesDL-FLAIR Images
    GM
        PSNR27.16 ± 0.54b,c34.03 ± 1.33b,c (+25.31%)
        NRMSE0.47 ± 0.039b,c0.33 ± 0.028b,c (−30.44%)
    WM
        PSNR29.51 ± 0.44b,c35.45 ± 1.48b,c (+20.14%)
        NRMSE0.37 ± 0.025b,c0.27 ± 0.024b,c (−27.32%)
    CSF
        PSNR34.83 ± 1.37b,c40.47 ± 1.37b,c (+16.21%)
        NRMSE0.29 ± 0.027b,c0.21 ± 0.017b,c (−27.36%)
    Aggregate intracranial tissues
        PSNR29.79 ± 0.71b35.90 ± 1.21b (+20.54%)
        NRMSE0.38 ± 0.024b0.27 ± 0.016b (−28.66%)
    • ↵a Values are means ± SD. Percentage changes in the PSNR and NRMSE for DL-FLAIR vs synthetic FLAIR images are in parentheses.

    • ↵b P < .001 for synthetic FLAIR vs DL-FLAIR images.

    • ↵c P < .001 for GM vs WM, GM vs CSF, and WM vs CSF.

PreviousNext
Back to top

In this issue

American Journal of Neuroradiology: 40 (2)
American Journal of Neuroradiology
Vol. 40, Issue 2
1 Feb 2019
  • Table of Contents
  • Index by author
  • Complete Issue (PDF)
Advertisement
Print
Download PDF
Email Article

Thank you for your interest in spreading the word on American Journal of Neuroradiology.

NOTE: We only request your email address so that the person you are recommending the page to knows that you wanted them to see it, and that it is not junk mail. We do not capture any email address.

Enter multiple addresses on separate lines or separate them with commas.
Improving the Quality of Synthetic FLAIR Images with Deep Learning Using a Conditional Generative Adversarial Network for Pixel-by-Pixel Image Translation
(Your Name) has sent you a message from American Journal of Neuroradiology
(Your Name) thought you would like to see the American Journal of Neuroradiology web site.
CAPTCHA
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Citation Tools
Improving the Quality of Synthetic FLAIR Images with Deep Learning Using a Conditional Generative Adversarial Network for Pixel-by-Pixel Image Translation
A. Hagiwara, Y. Otsuka, M. Hori, Y. Tachibana, K. Yokoyama, S. Fujita, C. Andica, K. Kamagata, R. Irie, S. Koshino, T. Maekawa, L. Chougar, A. Wada, M.Y. Takemura, N. Hattori, S. Aoki
American Journal of Neuroradiology Feb 2019, 40 (2) 224-230; DOI: 10.3174/ajnr.A5927

Citation Manager Formats

  • BibTeX
  • Bookends
  • EasyBib
  • EndNote (tagged)
  • EndNote 8 (xml)
  • Medlars
  • Mendeley
  • Papers
  • RefWorks Tagged
  • Ref Manager
  • RIS
  • Zotero
Share
Improving the Quality of Synthetic FLAIR Images with Deep Learning Using a Conditional Generative Adversarial Network for Pixel-by-Pixel Image Translation
A. Hagiwara, Y. Otsuka, M. Hori, Y. Tachibana, K. Yokoyama, S. Fujita, C. Andica, K. Kamagata, R. Irie, S. Koshino, T. Maekawa, L. Chougar, A. Wada, M.Y. Takemura, N. Hattori, S. Aoki
American Journal of Neuroradiology Feb 2019, 40 (2) 224-230; DOI: 10.3174/ajnr.A5927
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo
  • Tweet Widget
  • Facebook Like
  • Google Plus One
Purchase

Jump to section

  • Article
    • Abstract
    • ABBREVIATIONS:
    • Materials and Methods
    • Results
    • Discussion
    • Conclusions
    • Footnotes
    • References
  • Figures & Data
  • Info & Metrics
  • References
  • PDF

Related Articles

  • No related articles found.
  • PubMed
  • Google Scholar

Cited By...

  • 3D Quantitative Synthetic MRI in the Evaluation of Multiple Sclerosis Lesions
  • Crossref
  • Google Scholar

This article has not yet been cited by articles in journals that are participating in Crossref Cited-by Linking.

More in this TOC Section

  • Anatomic and Embryologic Analysis of the Dural Branches of the Ophthalmic Artery
  • Automated Cerebral Hemorrhage Detection Using RAPID
  • Analysis of Stroke Detection during the COVID-19 Pandemic Using Natural Language Processing of Radiology Reports
Show more ADULT BRAIN

Similar Articles

Advertisement

News and Updates

  • Lucien Levy Best Research Article Award
  • Thanks to our 2020 Distinguished Reviewers
  • Press Releases

Resources

  • Evidence-Based Medicine Level Guide
  • How to Participate in a Tweet Chat
  • AJNR Podcast Archive
  • Ideas for Publicizing Your Research
  • Librarian Resources
  • Terms and Conditions

Opportunities

  • Share Your Art in Perspectives
  • Get Peer Review Credit from Publons
  • Moderate a Tweet Chat

American Society of Neuroradiology

  • Neurographics
  • ASNR Annual Meeting
  • Fellowship Portal
  • Position Statements

© 2021 by the American Society of Neuroradiology | Print ISSN: 0195-6108 Online ISSN: 1936-959X

Powered by HighWire