Novel compressive sensing MRI methods with combined sparsifying transforms

Ying Dong, Jim Ji

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

3 Citations (Scopus)

Abstract

Compressive sensing (CS) is an emerging technique for fast MRI, which relies on the sparsity constraint of the underlying image to reduce the data acquisition requirement. Sparsifying transforms, such as total variation (TV), wavelet, curvelet, have been used in CS-MRI as regularization terms. Linear weighted summations of these regularization terms have also been used and tested. However, tuning the weights for individual terms is complicated and time-consuming. In this paper, a novel method that uses combined sparsifying transforms is proposed. This method applies transforms sequentially. It can avoid the artifacts associated with a single transform, as well as save the time of tuning the weights. Simulated results using in-vivo data show that the proposed method is efficient while providing similar or improved reconstruction quality.

Original languageEnglish
Title of host publicationProceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012
Pages721-724
Number of pages4
DOIs
Publication statusPublished - 2012
Externally publishedYes
EventIEEE-EMBS International Conference on Biomedical and Health Informatics, BHI 2012. In Conj. with the 8th Int. Symp.on Medical Devices and Biosensors and the 7th Int. Symp. on Biomedical and Health Engineering - Hong Kong and Shenzhen, China
Duration: 2 Jan 20127 Jan 2012

Other

OtherIEEE-EMBS International Conference on Biomedical and Health Informatics, BHI 2012. In Conj. with the 8th Int. Symp.on Medical Devices and Biosensors and the 7th Int. Symp. on Biomedical and Health Engineering
CountryChina
CityHong Kong and Shenzhen
Period2/1/127/1/12

Fingerprint

Magnetic resonance imaging
Tuning
Weights and Measures
Data acquisition
Artifacts

ASJC Scopus subject areas

  • Biomedical Engineering
  • Health Informatics

Cite this

Dong, Y., & Ji, J. (2012). Novel compressive sensing MRI methods with combined sparsifying transforms. In Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012 (pp. 721-724). [6211684] https://doi.org/10.1109/BHI.2012.6211684

Novel compressive sensing MRI methods with combined sparsifying transforms. / Dong, Ying; Ji, Jim.

Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012. 2012. p. 721-724 6211684.

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

Dong, Y & Ji, J 2012, Novel compressive sensing MRI methods with combined sparsifying transforms. in Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012., 6211684, pp. 721-724, IEEE-EMBS International Conference on Biomedical and Health Informatics, BHI 2012. In Conj. with the 8th Int. Symp.on Medical Devices and Biosensors and the 7th Int. Symp. on Biomedical and Health Engineering, Hong Kong and Shenzhen, China, 2/1/12. https://doi.org/10.1109/BHI.2012.6211684
Dong Y, Ji J. Novel compressive sensing MRI methods with combined sparsifying transforms. In Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012. 2012. p. 721-724. 6211684 https://doi.org/10.1109/BHI.2012.6211684
Dong, Ying ; Ji, Jim. / Novel compressive sensing MRI methods with combined sparsifying transforms. Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012. 2012. pp. 721-724
@inproceedings{a8213ced93a9440dbfe751f5544fd624,
title = "Novel compressive sensing MRI methods with combined sparsifying transforms",
abstract = "Compressive sensing (CS) is an emerging technique for fast MRI, which relies on the sparsity constraint of the underlying image to reduce the data acquisition requirement. Sparsifying transforms, such as total variation (TV), wavelet, curvelet, have been used in CS-MRI as regularization terms. Linear weighted summations of these regularization terms have also been used and tested. However, tuning the weights for individual terms is complicated and time-consuming. In this paper, a novel method that uses combined sparsifying transforms is proposed. This method applies transforms sequentially. It can avoid the artifacts associated with a single transform, as well as save the time of tuning the weights. Simulated results using in-vivo data show that the proposed method is efficient while providing similar or improved reconstruction quality.",
author = "Ying Dong and Jim Ji",
year = "2012",
doi = "10.1109/BHI.2012.6211684",
language = "English",
isbn = "9781457721779",
pages = "721--724",
booktitle = "Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012",

}

TY - GEN

T1 - Novel compressive sensing MRI methods with combined sparsifying transforms

AU - Dong, Ying

AU - Ji, Jim

PY - 2012

Y1 - 2012

N2 - Compressive sensing (CS) is an emerging technique for fast MRI, which relies on the sparsity constraint of the underlying image to reduce the data acquisition requirement. Sparsifying transforms, such as total variation (TV), wavelet, curvelet, have been used in CS-MRI as regularization terms. Linear weighted summations of these regularization terms have also been used and tested. However, tuning the weights for individual terms is complicated and time-consuming. In this paper, a novel method that uses combined sparsifying transforms is proposed. This method applies transforms sequentially. It can avoid the artifacts associated with a single transform, as well as save the time of tuning the weights. Simulated results using in-vivo data show that the proposed method is efficient while providing similar or improved reconstruction quality.

AB - Compressive sensing (CS) is an emerging technique for fast MRI, which relies on the sparsity constraint of the underlying image to reduce the data acquisition requirement. Sparsifying transforms, such as total variation (TV), wavelet, curvelet, have been used in CS-MRI as regularization terms. Linear weighted summations of these regularization terms have also been used and tested. However, tuning the weights for individual terms is complicated and time-consuming. In this paper, a novel method that uses combined sparsifying transforms is proposed. This method applies transforms sequentially. It can avoid the artifacts associated with a single transform, as well as save the time of tuning the weights. Simulated results using in-vivo data show that the proposed method is efficient while providing similar or improved reconstruction quality.

UR - http://www.scopus.com/inward/record.url?scp=84864208642&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864208642&partnerID=8YFLogxK

U2 - 10.1109/BHI.2012.6211684

DO - 10.1109/BHI.2012.6211684

M3 - Conference contribution

AN - SCOPUS:84864208642

SN - 9781457721779

SP - 721

EP - 724

BT - Proceedings - IEEE-EMBS International Conference on Biomedical and Health Informatics: Global Grand Challenge of Health Informatics, BHI 2012

ER -