Highly undersampled MRI using adaptive sparse representations

Saiprasad Ravishankar, Yoram Bresler

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

Abstract

Compressed sensing (CS) exploits the sparsity of MR images to enable accurate reconstruction from undersampled k-space data. Recent CS methods have employed analytical sparsifying transforms such as wavelets and finite differences. In this paper, we propose a novel framework for adaptively learning the sparsifying transform (dictionary), and reconstructing the image simultaneously from highly undersampled k-space data. The sparsity is enforced on overlapping image patches. The proposed alternating reconstruction algorithm learns the sparsifying dictionary, and uses it to remove aliasing and noise in one step, and subsequently restores and fills-in the k-space data in the other step. Experimental results demonstrate dramatic improvements in reconstruction error using the proposed adaptive dictionary as compared to previous CS methods.

Original languageEnglish (US)
Title of host publication2011 8th IEEE International Symposium on Biomedical Imaging
Subtitle of host publicationFrom Nano to Macro, ISBI'11
Pages1585-1588
Number of pages4
DOIs
StatePublished - Nov 2 2011
Event2011 8th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI'11 - Chicago, IL, United States
Duration: Mar 30 2011Apr 2 2011

Publication series

NameProceedings - International Symposium on Biomedical Imaging
ISSN (Print)1945-7928
ISSN (Electronic)1945-8452

Other

Other2011 8th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI'11
CountryUnited States
CityChicago, IL
Period3/30/114/2/11

Fingerprint

Compressed sensing
Glossaries
Magnetic resonance imaging
Noise
Learning

Keywords

  • Compressed sensing
  • dictionary learning
  • Image reconstruction
  • Magnetic resonance imaging

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Ravishankar, S., & Bresler, Y. (2011). Highly undersampled MRI using adaptive sparse representations. In 2011 8th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI'11 (pp. 1585-1588). [5872705] (Proceedings - International Symposium on Biomedical Imaging). https://doi.org/10.1109/ISBI.2011.5872705

Highly undersampled MRI using adaptive sparse representations. / Ravishankar, Saiprasad; Bresler, Yoram.

2011 8th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI'11. 2011. p. 1585-1588 5872705 (Proceedings - International Symposium on Biomedical Imaging).

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

Ravishankar, S & Bresler, Y 2011, Highly undersampled MRI using adaptive sparse representations. in 2011 8th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI'11., 5872705, Proceedings - International Symposium on Biomedical Imaging, pp. 1585-1588, 2011 8th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI'11, Chicago, IL, United States, 3/30/11. https://doi.org/10.1109/ISBI.2011.5872705
Ravishankar S, Bresler Y. Highly undersampled MRI using adaptive sparse representations. In 2011 8th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI'11. 2011. p. 1585-1588. 5872705. (Proceedings - International Symposium on Biomedical Imaging). https://doi.org/10.1109/ISBI.2011.5872705
Ravishankar, Saiprasad ; Bresler, Yoram. / Highly undersampled MRI using adaptive sparse representations. 2011 8th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI'11. 2011. pp. 1585-1588 (Proceedings - International Symposium on Biomedical Imaging).
@inproceedings{238f391ad0854dd688b6c645d6528843,
title = "Highly undersampled MRI using adaptive sparse representations",
abstract = "Compressed sensing (CS) exploits the sparsity of MR images to enable accurate reconstruction from undersampled k-space data. Recent CS methods have employed analytical sparsifying transforms such as wavelets and finite differences. In this paper, we propose a novel framework for adaptively learning the sparsifying transform (dictionary), and reconstructing the image simultaneously from highly undersampled k-space data. The sparsity is enforced on overlapping image patches. The proposed alternating reconstruction algorithm learns the sparsifying dictionary, and uses it to remove aliasing and noise in one step, and subsequently restores and fills-in the k-space data in the other step. Experimental results demonstrate dramatic improvements in reconstruction error using the proposed adaptive dictionary as compared to previous CS methods.",
keywords = "Compressed sensing, dictionary learning, Image reconstruction, Magnetic resonance imaging",
author = "Saiprasad Ravishankar and Yoram Bresler",
year = "2011",
month = "11",
day = "2",
doi = "10.1109/ISBI.2011.5872705",
language = "English (US)",
isbn = "9781424441280",
series = "Proceedings - International Symposium on Biomedical Imaging",
pages = "1585--1588",
booktitle = "2011 8th IEEE International Symposium on Biomedical Imaging",

}

TY - GEN

T1 - Highly undersampled MRI using adaptive sparse representations

AU - Ravishankar, Saiprasad

AU - Bresler, Yoram

PY - 2011/11/2

Y1 - 2011/11/2

N2 - Compressed sensing (CS) exploits the sparsity of MR images to enable accurate reconstruction from undersampled k-space data. Recent CS methods have employed analytical sparsifying transforms such as wavelets and finite differences. In this paper, we propose a novel framework for adaptively learning the sparsifying transform (dictionary), and reconstructing the image simultaneously from highly undersampled k-space data. The sparsity is enforced on overlapping image patches. The proposed alternating reconstruction algorithm learns the sparsifying dictionary, and uses it to remove aliasing and noise in one step, and subsequently restores and fills-in the k-space data in the other step. Experimental results demonstrate dramatic improvements in reconstruction error using the proposed adaptive dictionary as compared to previous CS methods.

AB - Compressed sensing (CS) exploits the sparsity of MR images to enable accurate reconstruction from undersampled k-space data. Recent CS methods have employed analytical sparsifying transforms such as wavelets and finite differences. In this paper, we propose a novel framework for adaptively learning the sparsifying transform (dictionary), and reconstructing the image simultaneously from highly undersampled k-space data. The sparsity is enforced on overlapping image patches. The proposed alternating reconstruction algorithm learns the sparsifying dictionary, and uses it to remove aliasing and noise in one step, and subsequently restores and fills-in the k-space data in the other step. Experimental results demonstrate dramatic improvements in reconstruction error using the proposed adaptive dictionary as compared to previous CS methods.

KW - Compressed sensing

KW - dictionary learning

KW - Image reconstruction

KW - Magnetic resonance imaging

UR - http://www.scopus.com/inward/record.url?scp=80055059617&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80055059617&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2011.5872705

DO - 10.1109/ISBI.2011.5872705

M3 - Conference contribution

AN - SCOPUS:80055059617

SN - 9781424441280

T3 - Proceedings - International Symposium on Biomedical Imaging

SP - 1585

EP - 1588

BT - 2011 8th IEEE International Symposium on Biomedical Imaging

ER -