Dictionary learning for compressive parameter mapping in magnetic resonance imaging

Benjamin P. Berman, Mahesh B. Keerthivasan, Zhitao Li, Diego R Martin, Maria I Altbach, Ali Bilgin

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

Abstract

Parameter mapping is a valuable quantitative tool for soft tissue contrast. Accelerated data acquisition is critical for clinical utility, which has lead to various novel reconstruction techniques. In this work, a model-based compressed sensing method is extended to include a sparse regularization that is learned from the principal component coefficient. The principal components for a range of T2 decay curves are computed, and the coefficients of the principal components are reconstructed. These coefficient maps share coherent spatial structures, suggesting a patch{based dictionary is a well suited sparse transformation. This transformation is learned from the coefficients themselves. The proposed reconstruction is suited for non-Cartesian, multi-channel data. The dictionary constraint leads to parameter maps with less noise and less aliasing for high amounts of acceleration.

Original languageEnglish (US)
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
PublisherSPIE
Volume9597
ISBN (Print)9781628417630, 9781628417630
DOIs
StatePublished - 2015
EventWavelets and Sparsity XVI - San Diego, United States
Duration: Aug 10 2015Aug 12 2015

Other

OtherWavelets and Sparsity XVI
CountryUnited States
CitySan Diego
Period8/10/158/12/15

Fingerprint

dictionaries
Magnetic Resonance Imaging
Magnetic resonance
Glossaries
learning
magnetic resonance
Principal Components
Imaging techniques
Compressed sensing
Coefficient
coefficients
Data acquisition
Tissue
Coherent Structures
Compressed Sensing
Aliasing
Soft Tissue
Spatial Structure
Data Acquisition
data acquisition

Keywords

  • Compressed sensing
  • MRI
  • Radial
  • Sparsity
  • T

ASJC Scopus subject areas

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

Cite this

Berman, B. P., Keerthivasan, M. B., Li, Z., Martin, D. R., Altbach, M. I., & Bilgin, A. (2015). Dictionary learning for compressive parameter mapping in magnetic resonance imaging. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 9597). [959707] SPIE. https://doi.org/10.1117/12.2187088

Dictionary learning for compressive parameter mapping in magnetic resonance imaging. / Berman, Benjamin P.; Keerthivasan, Mahesh B.; Li, Zhitao; Martin, Diego R; Altbach, Maria I; Bilgin, Ali.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 9597 SPIE, 2015. 959707.

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

Berman, BP, Keerthivasan, MB, Li, Z, Martin, DR, Altbach, MI & Bilgin, A 2015, Dictionary learning for compressive parameter mapping in magnetic resonance imaging. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 9597, 959707, SPIE, Wavelets and Sparsity XVI, San Diego, United States, 8/10/15. https://doi.org/10.1117/12.2187088
Berman BP, Keerthivasan MB, Li Z, Martin DR, Altbach MI, Bilgin A. Dictionary learning for compressive parameter mapping in magnetic resonance imaging. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 9597. SPIE. 2015. 959707 https://doi.org/10.1117/12.2187088
Berman, Benjamin P. ; Keerthivasan, Mahesh B. ; Li, Zhitao ; Martin, Diego R ; Altbach, Maria I ; Bilgin, Ali. / Dictionary learning for compressive parameter mapping in magnetic resonance imaging. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 9597 SPIE, 2015.
@inproceedings{8a1a965adac14c859b08493f84c11f16,
title = "Dictionary learning for compressive parameter mapping in magnetic resonance imaging",
abstract = "Parameter mapping is a valuable quantitative tool for soft tissue contrast. Accelerated data acquisition is critical for clinical utility, which has lead to various novel reconstruction techniques. In this work, a model-based compressed sensing method is extended to include a sparse regularization that is learned from the principal component coefficient. The principal components for a range of T2 decay curves are computed, and the coefficients of the principal components are reconstructed. These coefficient maps share coherent spatial structures, suggesting a patch{based dictionary is a well suited sparse transformation. This transformation is learned from the coefficients themselves. The proposed reconstruction is suited for non-Cartesian, multi-channel data. The dictionary constraint leads to parameter maps with less noise and less aliasing for high amounts of acceleration.",
keywords = "Compressed sensing, MRI, Radial, Sparsity, T",
author = "Berman, {Benjamin P.} and Keerthivasan, {Mahesh B.} and Zhitao Li and Martin, {Diego R} and Altbach, {Maria I} and Ali Bilgin",
year = "2015",
doi = "10.1117/12.2187088",
language = "English (US)",
isbn = "9781628417630",
volume = "9597",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",
publisher = "SPIE",

}

TY - GEN

T1 - Dictionary learning for compressive parameter mapping in magnetic resonance imaging

AU - Berman, Benjamin P.

AU - Keerthivasan, Mahesh B.

AU - Li, Zhitao

AU - Martin, Diego R

AU - Altbach, Maria I

AU - Bilgin, Ali

PY - 2015

Y1 - 2015

N2 - Parameter mapping is a valuable quantitative tool for soft tissue contrast. Accelerated data acquisition is critical for clinical utility, which has lead to various novel reconstruction techniques. In this work, a model-based compressed sensing method is extended to include a sparse regularization that is learned from the principal component coefficient. The principal components for a range of T2 decay curves are computed, and the coefficients of the principal components are reconstructed. These coefficient maps share coherent spatial structures, suggesting a patch{based dictionary is a well suited sparse transformation. This transformation is learned from the coefficients themselves. The proposed reconstruction is suited for non-Cartesian, multi-channel data. The dictionary constraint leads to parameter maps with less noise and less aliasing for high amounts of acceleration.

AB - Parameter mapping is a valuable quantitative tool for soft tissue contrast. Accelerated data acquisition is critical for clinical utility, which has lead to various novel reconstruction techniques. In this work, a model-based compressed sensing method is extended to include a sparse regularization that is learned from the principal component coefficient. The principal components for a range of T2 decay curves are computed, and the coefficients of the principal components are reconstructed. These coefficient maps share coherent spatial structures, suggesting a patch{based dictionary is a well suited sparse transformation. This transformation is learned from the coefficients themselves. The proposed reconstruction is suited for non-Cartesian, multi-channel data. The dictionary constraint leads to parameter maps with less noise and less aliasing for high amounts of acceleration.

KW - Compressed sensing

KW - MRI

KW - Radial

KW - Sparsity

KW - T

UR - http://www.scopus.com/inward/record.url?scp=84951310799&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84951310799&partnerID=8YFLogxK

U2 - 10.1117/12.2187088

DO - 10.1117/12.2187088

M3 - Conference contribution

SN - 9781628417630

SN - 9781628417630

VL - 9597

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

PB - SPIE

ER -