Image quality transfer and applications in diffusion MRI

Publication: Research - peer-reviewJournal article – Annual report year: 2017

Documents

DOI

  • Author: Alexander, Daniel C.

    University College London

  • Author: Zikic, Darko

    Microsoft USA

  • Author: Ghosh, Aurobrata

    University College London

  • Author: Tanno, Ryutaro

    University College London

  • Author: Wottschel, Viktor

    University College London

  • Author: Zhang, Jiaying

    University College London

  • Author: Kaden, Enrico

    University College London

  • Author: Dyrby, Tim B.

    Image Analysis & Computer Graphics, Department of Applied Mathematics and Computer Science , Technical University of Denmark, Richard Petersens Plads, 2800, Kgs. Lyngby, Denmark

  • Author: Sotiropoulos, Stamatios N.

    University of Oxford

  • Author: Zhang, Hui

    University College London

  • Author: Criminisi, Antonio

    Microsoft USA

View graph of relations

This paper introduces a new computational imaging technique called image quality transfer (IQT). IQT uses machine learning to transfer the rich information available from one-off experimental medical imaging devices to the abundant but lower-quality data from routine acquisitions. The procedure uses matched pairs to learn mappings from low-quality to corresponding high-quality images. Once learned, these mappings then augment unseen low quality images, for example by enhancing image resolution or information content. Here, we demonstrate IQT using a simple patch-regression implementation and the uniquely rich diffusion MRI data set from the human connectome project (HCP). Results highlight potential benefits of IQT in both brain connectivity mapping and microstructure imaging. In brain connectivity mapping, IQT reveals, from standard data sets, thin connection pathways that tractography normally requires specialised data to reconstruct. In microstructure imaging, IQT shows potential in estimating, from standard “single-shell” data (one non-zero b-value), maps of microstructural parameters that normally require specialised multi-shell data. Further experiments show strong generalisability, highlighting IQT's benefits even when the training set does not directly represent the application domain. The concept extends naturally to many other imaging modalities and reconstruction problems.

Original languageEnglish
JournalNeuroImage
Volume152
Pages (from-to)283-298
Number of pages16
ISSN1053-8119
DOIs
StatePublished - 15 May 2017
CitationsWeb of Science® Times Cited: 4
Download as:
Download as PDF
Select render style:
APAAuthorCBE/CSEHarvardMLAStandardVancouverShortLong
PDF
Download as HTML
Select render style:
APAAuthorCBE/CSEHarvardMLAStandardVancouverShortLong
HTML
Download as Word
Select render style:
APAAuthorCBE/CSEHarvardMLAStandardVancouverShortLong
Word

ID: 142114787