Parallel magnetic resonance imaging reconstruction algorithm by 3-dimension directional Haar tight framelet regularization

Yan-Ran Li, Xiaosheng Zhuang*

*Corresponding author for this work

Research output: Chapters, Conference Papers, Creative and Literary WorksRGC 32 - Refereed conference paper (with host publication)peer-review

5 Citations (Scopus)

Abstract

In this paper, a 3-dimension directional Haar tight framelet (3DHF) is used to detect the related features between coil images in parallel magnetic resonance imaging (pMRI). Such a Haar tight framelet has an extremely simple geometric structure in the sense that all the high-pass filters in its underlying filter bank have only two nonzero coefficients with opposite signs. A pMRI optimization model, which we coined 3DHF-SPIRiT, by regularizing the 3DHF features on the 3-D coil image data is proposed to reduce the aliasing artifacts caused by the downsampling operation in the k-space (Fourier) domain, which can be solved by alternating direction method of multipliers (ADMM) scheme. Numerical experiments are provided to demonstrate the superiority and efficiency of our 3DHF-SPIRiT model.
Original languageEnglish
Title of host publicationWavelets and Sparsity XVIII
EditorsDimitri Van De Ville, Manos Papadakis, Yue M. Lu
PublisherSPIE
ISBN (Electronic)9781510629707
ISBN (Print)9781510629691
DOIs
Publication statusPublished - Aug 2019
EventConference on Wavelets and Sparsity XVIII - San Diego, United States
Duration: 13 Aug 201915 Aug 2019

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume11138
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

ConferenceConference on Wavelets and Sparsity XVIII
PlaceUnited States
CitySan Diego
Period13/08/1915/08/19

Research Keywords

  • 3-D regularization
  • ADMM
  • directional Haar tight framelets
  • GRAPPA
  • pMRI
  • SENSE
  • SPIRiT

Fingerprint

Dive into the research topics of 'Parallel magnetic resonance imaging reconstruction algorithm by 3-dimension directional Haar tight framelet regularization'. Together they form a unique fingerprint.

Cite this