Sparse decomposition learning based dynamic MRI reconstruction

Peifei Zhu, Qieshi Zhang, Seiichiro Kamata

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Dynamic MRI is widely used for many clinical exams but slow data acquisition becomes a serious problem. The application of Compressed Sensing (CS) demonstrated great potential to increase imaging speed. However, the performance of CS is largely depending on the sparsity of image sequence in the transform domain, where there are still a lot to be improved. In this work, the sparsity is exploited by proposed Sparse Decomposition Learning (SDL) algorithm, which is a combination of low-rank plus sparsity and Blind Compressed Sensing (BCS). With this decomposition, only sparsity component is modeled as a sparse linear combination of temporal basis functions. This enables coefficients to be sparser and remain more details of dynamic components comparing learning the whole images. A reconstruction is performed on the undersampled data where joint multicoil data consistency is enforced by combing Parallel Imaging (PI). The experimental results show the proposed methods decrease about 15∼20% of Mean Square Error (MSE) compared to other existing methods.

Original languageEnglish
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
PublisherSPIE
Volume9445
ISBN (Print)9781628415605
DOIs
Publication statusPublished - 2015
Event7th International Conference on Machine Vision, ICMV 2014 - Milan
Duration: 2014 Nov 192014 Nov 21

Other

Other7th International Conference on Machine Vision, ICMV 2014
CityMilan
Period14/11/1914/11/21

Fingerprint

Compressed sensing
Sparsity
Magnetic resonance imaging
learning
Compressed Sensing
Decomposition
decomposition
Decompose
Imaging techniques
Imaging
Data Consistency
Mean square error
Learning algorithms
data acquisition
Data acquisition
Decomposition Algorithm
Image Sequence
Data Acquisition
Basis Functions
Linear Combination

Keywords

  • Compressed Sensing (CS)
  • Dynamic MRI
  • Parallel imaging (PI)
  • Sparse Decomposition Learning (SDL)

ASJC Scopus subject areas

  • Applied Mathematics
  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics

Cite this

Zhu, P., Zhang, Q., & Kamata, S. (2015). Sparse decomposition learning based dynamic MRI reconstruction. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 9445). [944516] SPIE. https://doi.org/10.1117/12.2180534

Sparse decomposition learning based dynamic MRI reconstruction. / Zhu, Peifei; Zhang, Qieshi; Kamata, Seiichiro.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 9445 SPIE, 2015. 944516.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhu, P, Zhang, Q & Kamata, S 2015, Sparse decomposition learning based dynamic MRI reconstruction. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 9445, 944516, SPIE, 7th International Conference on Machine Vision, ICMV 2014, Milan, 14/11/19. https://doi.org/10.1117/12.2180534
Zhu P, Zhang Q, Kamata S. Sparse decomposition learning based dynamic MRI reconstruction. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 9445. SPIE. 2015. 944516 https://doi.org/10.1117/12.2180534
Zhu, Peifei ; Zhang, Qieshi ; Kamata, Seiichiro. / Sparse decomposition learning based dynamic MRI reconstruction. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 9445 SPIE, 2015.
@inproceedings{5c2ca69a0412489eb8a131d5f71d92e7,
title = "Sparse decomposition learning based dynamic MRI reconstruction",
abstract = "Dynamic MRI is widely used for many clinical exams but slow data acquisition becomes a serious problem. The application of Compressed Sensing (CS) demonstrated great potential to increase imaging speed. However, the performance of CS is largely depending on the sparsity of image sequence in the transform domain, where there are still a lot to be improved. In this work, the sparsity is exploited by proposed Sparse Decomposition Learning (SDL) algorithm, which is a combination of low-rank plus sparsity and Blind Compressed Sensing (BCS). With this decomposition, only sparsity component is modeled as a sparse linear combination of temporal basis functions. This enables coefficients to be sparser and remain more details of dynamic components comparing learning the whole images. A reconstruction is performed on the undersampled data where joint multicoil data consistency is enforced by combing Parallel Imaging (PI). The experimental results show the proposed methods decrease about 15∼20{\%} of Mean Square Error (MSE) compared to other existing methods.",
keywords = "Compressed Sensing (CS), Dynamic MRI, Parallel imaging (PI), Sparse Decomposition Learning (SDL)",
author = "Peifei Zhu and Qieshi Zhang and Seiichiro Kamata",
year = "2015",
doi = "10.1117/12.2180534",
language = "English",
isbn = "9781628415605",
volume = "9445",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",
publisher = "SPIE",

}

TY - GEN

T1 - Sparse decomposition learning based dynamic MRI reconstruction

AU - Zhu, Peifei

AU - Zhang, Qieshi

AU - Kamata, Seiichiro

PY - 2015

Y1 - 2015

N2 - Dynamic MRI is widely used for many clinical exams but slow data acquisition becomes a serious problem. The application of Compressed Sensing (CS) demonstrated great potential to increase imaging speed. However, the performance of CS is largely depending on the sparsity of image sequence in the transform domain, where there are still a lot to be improved. In this work, the sparsity is exploited by proposed Sparse Decomposition Learning (SDL) algorithm, which is a combination of low-rank plus sparsity and Blind Compressed Sensing (BCS). With this decomposition, only sparsity component is modeled as a sparse linear combination of temporal basis functions. This enables coefficients to be sparser and remain more details of dynamic components comparing learning the whole images. A reconstruction is performed on the undersampled data where joint multicoil data consistency is enforced by combing Parallel Imaging (PI). The experimental results show the proposed methods decrease about 15∼20% of Mean Square Error (MSE) compared to other existing methods.

AB - Dynamic MRI is widely used for many clinical exams but slow data acquisition becomes a serious problem. The application of Compressed Sensing (CS) demonstrated great potential to increase imaging speed. However, the performance of CS is largely depending on the sparsity of image sequence in the transform domain, where there are still a lot to be improved. In this work, the sparsity is exploited by proposed Sparse Decomposition Learning (SDL) algorithm, which is a combination of low-rank plus sparsity and Blind Compressed Sensing (BCS). With this decomposition, only sparsity component is modeled as a sparse linear combination of temporal basis functions. This enables coefficients to be sparser and remain more details of dynamic components comparing learning the whole images. A reconstruction is performed on the undersampled data where joint multicoil data consistency is enforced by combing Parallel Imaging (PI). The experimental results show the proposed methods decrease about 15∼20% of Mean Square Error (MSE) compared to other existing methods.

KW - Compressed Sensing (CS)

KW - Dynamic MRI

KW - Parallel imaging (PI)

KW - Sparse Decomposition Learning (SDL)

UR - http://www.scopus.com/inward/record.url?scp=84924347653&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84924347653&partnerID=8YFLogxK

U2 - 10.1117/12.2180534

DO - 10.1117/12.2180534

M3 - Conference contribution

AN - SCOPUS:84924347653

SN - 9781628415605

VL - 9445

BT - Proceedings of SPIE - The International Society for Optical Engineering

PB - SPIE

ER -