Compressed sensing MRI using double sparsity with additional training images

Chenmin Tang, Norihito Inamuro, Takashi Ijiri, Akira Hirabayashi

Research output: ResearchConference contribution

Abstract

The compressed sensing using dictionary learning has led to state-of-the-art results for magnetic resonance imaging (MRI) reconstruction from highly under-sampled measurements. Dictionary learning had been considered time-consuming especially when the patch size or the number of training patches is large. Recently, double sparsity model and online dictionary learning algorithm were proposed to obtain dictionaries with much less computational time. In this paper, we propose an efficient MRI reconstruction method by adopting the double sparsity model with the online dictionary learning method. Besides, for better reconstruction, we use separately prepared fully-sampled MRI images to train dictionaries. We compare results of the proposed technique to traditional offline methods with and without double sparsity model. Our simulation results show that the proposed technique is approximately twice faster than the traditional methods while maintaining the same reconstruction quality. Furthermore, our technique performed even better for lower sampling rate.

LanguageEnglish
Title of host publication2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages801-805
Number of pages5
ISBN (Electronic)9781509041176
DOIs
StatePublished - 2017 Jun 16
Externally publishedYes
Event2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - New Orleans, United States
Duration: 2017 Mar 52017 Mar 9

Other

Other2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017
CountryUnited States
CityNew Orleans
Period17/3/517/3/9

Fingerprint

Compressed sensing
Magnetic resonance
Glossaries
Imaging techniques
Learning algorithms
Sampling

Keywords

  • compressed sensing
  • double sparsity model
  • MRI
  • online dictionary learning

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Tang, C., Inamuro, N., Ijiri, T., & Hirabayashi, A. (2017). Compressed sensing MRI using double sparsity with additional training images. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings (pp. 801-805). [7952266] Institute of Electrical and Electronics Engineers Inc.. DOI: 10.1109/ICASSP.2017.7952266

Compressed sensing MRI using double sparsity with additional training images. / Tang, Chenmin; Inamuro, Norihito; Ijiri, Takashi; Hirabayashi, Akira.

2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 801-805 7952266.

Research output: ResearchConference contribution

Tang, C, Inamuro, N, Ijiri, T & Hirabayashi, A 2017, Compressed sensing MRI using double sparsity with additional training images. in 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings., 7952266, Institute of Electrical and Electronics Engineers Inc., pp. 801-805, 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017, New Orleans, United States, 17/3/5. DOI: 10.1109/ICASSP.2017.7952266
Tang C, Inamuro N, Ijiri T, Hirabayashi A. Compressed sensing MRI using double sparsity with additional training images. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc.2017. p. 801-805. 7952266. Available from, DOI: 10.1109/ICASSP.2017.7952266
Tang, Chenmin ; Inamuro, Norihito ; Ijiri, Takashi ; Hirabayashi, Akira. / Compressed sensing MRI using double sparsity with additional training images. 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 801-805
@inbook{82c55319c5fb46b182c16afeed431ee7,
title = "Compressed sensing MRI using double sparsity with additional training images",
abstract = "The compressed sensing using dictionary learning has led to state-of-the-art results for magnetic resonance imaging (MRI) reconstruction from highly under-sampled measurements. Dictionary learning had been considered time-consuming especially when the patch size or the number of training patches is large. Recently, double sparsity model and online dictionary learning algorithm were proposed to obtain dictionaries with much less computational time. In this paper, we propose an efficient MRI reconstruction method by adopting the double sparsity model with the online dictionary learning method. Besides, for better reconstruction, we use separately prepared fully-sampled MRI images to train dictionaries. We compare results of the proposed technique to traditional offline methods with and without double sparsity model. Our simulation results show that the proposed technique is approximately twice faster than the traditional methods while maintaining the same reconstruction quality. Furthermore, our technique performed even better for lower sampling rate.",
keywords = "compressed sensing, double sparsity model, MRI, online dictionary learning",
author = "Chenmin Tang and Norihito Inamuro and Takashi Ijiri and Akira Hirabayashi",
year = "2017",
month = "6",
doi = "10.1109/ICASSP.2017.7952266",
pages = "801--805",
booktitle = "2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

TY - CHAP

T1 - Compressed sensing MRI using double sparsity with additional training images

AU - Tang,Chenmin

AU - Inamuro,Norihito

AU - Ijiri,Takashi

AU - Hirabayashi,Akira

PY - 2017/6/16

Y1 - 2017/6/16

N2 - The compressed sensing using dictionary learning has led to state-of-the-art results for magnetic resonance imaging (MRI) reconstruction from highly under-sampled measurements. Dictionary learning had been considered time-consuming especially when the patch size or the number of training patches is large. Recently, double sparsity model and online dictionary learning algorithm were proposed to obtain dictionaries with much less computational time. In this paper, we propose an efficient MRI reconstruction method by adopting the double sparsity model with the online dictionary learning method. Besides, for better reconstruction, we use separately prepared fully-sampled MRI images to train dictionaries. We compare results of the proposed technique to traditional offline methods with and without double sparsity model. Our simulation results show that the proposed technique is approximately twice faster than the traditional methods while maintaining the same reconstruction quality. Furthermore, our technique performed even better for lower sampling rate.

AB - The compressed sensing using dictionary learning has led to state-of-the-art results for magnetic resonance imaging (MRI) reconstruction from highly under-sampled measurements. Dictionary learning had been considered time-consuming especially when the patch size or the number of training patches is large. Recently, double sparsity model and online dictionary learning algorithm were proposed to obtain dictionaries with much less computational time. In this paper, we propose an efficient MRI reconstruction method by adopting the double sparsity model with the online dictionary learning method. Besides, for better reconstruction, we use separately prepared fully-sampled MRI images to train dictionaries. We compare results of the proposed technique to traditional offline methods with and without double sparsity model. Our simulation results show that the proposed technique is approximately twice faster than the traditional methods while maintaining the same reconstruction quality. Furthermore, our technique performed even better for lower sampling rate.

KW - compressed sensing

KW - double sparsity model

KW - MRI

KW - online dictionary learning

UR - http://www.scopus.com/inward/record.url?scp=85023766734&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85023766734&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2017.7952266

DO - 10.1109/ICASSP.2017.7952266

M3 - Conference contribution

SP - 801

EP - 805

BT - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -