Using GPUs to accelerate advanced MRI reconstruction with field inhomogeneity compensation

Yue Zhuo, Xiao Long Wu, Justin P. Haldar, Thibault Marin, Wen mei W Hwu, Zhi Pei Liang, Bradley P. Sutton

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

This chapter focuses on a GPU implementation for a fast advanced non-Cartesian MRI reconstruction algorithm with field inhomogeneity compensation. Magnetic resonance imaging (MRI) is a flexible diagnostic tool, providing image contrast relating to the structure, function, and biochemistry of virtually every system in the body. However, the technique is generally slow and has low sensitivity, which limits its application in the clinical environment. Several challenges exist that limit the application of MRI in the clinical environment. Traditionally, the main limitations in MRI have been due to the manner in which data are sampled in clinical scans. The techniques of tiling have been applied with constant memory, loop invariant code motion, storing variables in registers, and using single-precision floating-point computations on the GPU kernels. The parallel structure of the reconstruction algorithms makes it suitable for parallel programming on GPUs. Accelerating this kind of algorithm can allow for more accurate image reconstruction while keeping computation times short enough for clinical use. Thus, the use of GPUs will enable improved trade-offs between data acquisition time, signal-to-noise ratio, and the severity of artifacts owing to nonideal physical effects during the MRI imaging experiment. © 2011

Original languageEnglish (US)
Title of host publicationGPU Computing Gems Emerald Edition
PublisherElsevier Inc.
Pages709-722
Number of pages14
ISBN (Print)9780123849885
DOIs
StatePublished - Dec 1 2011

Fingerprint

Magnetic resonance imaging
Biochemistry
Parallel programming
Image reconstruction
Data acquisition
Signal to noise ratio
Graphics processing unit
Compensation and Redress
Imaging techniques
Data storage equipment
Experiments

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Zhuo, Y., Wu, X. L., Haldar, J. P., Marin, T., Hwu, W. M. W., Liang, Z. P., & Sutton, B. P. (2011). Using GPUs to accelerate advanced MRI reconstruction with field inhomogeneity compensation. In GPU Computing Gems Emerald Edition (pp. 709-722). Elsevier Inc.. https://doi.org/10.1016/B978-0-12-384988-5.00044-9

Using GPUs to accelerate advanced MRI reconstruction with field inhomogeneity compensation. / Zhuo, Yue; Wu, Xiao Long; Haldar, Justin P.; Marin, Thibault; Hwu, Wen mei W; Liang, Zhi Pei; Sutton, Bradley P.

GPU Computing Gems Emerald Edition. Elsevier Inc., 2011. p. 709-722.

Research output: Chapter in Book/Report/Conference proceedingChapter

Zhuo Y, Wu XL, Haldar JP, Marin T, Hwu WMW, Liang ZP et al. Using GPUs to accelerate advanced MRI reconstruction with field inhomogeneity compensation. In GPU Computing Gems Emerald Edition. Elsevier Inc. 2011. p. 709-722 https://doi.org/10.1016/B978-0-12-384988-5.00044-9
Zhuo, Yue ; Wu, Xiao Long ; Haldar, Justin P. ; Marin, Thibault ; Hwu, Wen mei W ; Liang, Zhi Pei ; Sutton, Bradley P. / Using GPUs to accelerate advanced MRI reconstruction with field inhomogeneity compensation. GPU Computing Gems Emerald Edition. Elsevier Inc., 2011. pp. 709-722
@inbook{0286f2b4ab644a7b8f4b286d69c020cd,
title = "Using GPUs to accelerate advanced MRI reconstruction with field inhomogeneity compensation",
abstract = "This chapter focuses on a GPU implementation for a fast advanced non-Cartesian MRI reconstruction algorithm with field inhomogeneity compensation. Magnetic resonance imaging (MRI) is a flexible diagnostic tool, providing image contrast relating to the structure, function, and biochemistry of virtually every system in the body. However, the technique is generally slow and has low sensitivity, which limits its application in the clinical environment. Several challenges exist that limit the application of MRI in the clinical environment. Traditionally, the main limitations in MRI have been due to the manner in which data are sampled in clinical scans. The techniques of tiling have been applied with constant memory, loop invariant code motion, storing variables in registers, and using single-precision floating-point computations on the GPU kernels. The parallel structure of the reconstruction algorithms makes it suitable for parallel programming on GPUs. Accelerating this kind of algorithm can allow for more accurate image reconstruction while keeping computation times short enough for clinical use. Thus, the use of GPUs will enable improved trade-offs between data acquisition time, signal-to-noise ratio, and the severity of artifacts owing to nonideal physical effects during the MRI imaging experiment. {\circledC} 2011",
author = "Yue Zhuo and Wu, {Xiao Long} and Haldar, {Justin P.} and Thibault Marin and Hwu, {Wen mei W} and Liang, {Zhi Pei} and Sutton, {Bradley P.}",
year = "2011",
month = "12",
day = "1",
doi = "10.1016/B978-0-12-384988-5.00044-9",
language = "English (US)",
isbn = "9780123849885",
pages = "709--722",
booktitle = "GPU Computing Gems Emerald Edition",
publisher = "Elsevier Inc.",

}

TY - CHAP

T1 - Using GPUs to accelerate advanced MRI reconstruction with field inhomogeneity compensation

AU - Zhuo, Yue

AU - Wu, Xiao Long

AU - Haldar, Justin P.

AU - Marin, Thibault

AU - Hwu, Wen mei W

AU - Liang, Zhi Pei

AU - Sutton, Bradley P.

PY - 2011/12/1

Y1 - 2011/12/1

N2 - This chapter focuses on a GPU implementation for a fast advanced non-Cartesian MRI reconstruction algorithm with field inhomogeneity compensation. Magnetic resonance imaging (MRI) is a flexible diagnostic tool, providing image contrast relating to the structure, function, and biochemistry of virtually every system in the body. However, the technique is generally slow and has low sensitivity, which limits its application in the clinical environment. Several challenges exist that limit the application of MRI in the clinical environment. Traditionally, the main limitations in MRI have been due to the manner in which data are sampled in clinical scans. The techniques of tiling have been applied with constant memory, loop invariant code motion, storing variables in registers, and using single-precision floating-point computations on the GPU kernels. The parallel structure of the reconstruction algorithms makes it suitable for parallel programming on GPUs. Accelerating this kind of algorithm can allow for more accurate image reconstruction while keeping computation times short enough for clinical use. Thus, the use of GPUs will enable improved trade-offs between data acquisition time, signal-to-noise ratio, and the severity of artifacts owing to nonideal physical effects during the MRI imaging experiment. © 2011

AB - This chapter focuses on a GPU implementation for a fast advanced non-Cartesian MRI reconstruction algorithm with field inhomogeneity compensation. Magnetic resonance imaging (MRI) is a flexible diagnostic tool, providing image contrast relating to the structure, function, and biochemistry of virtually every system in the body. However, the technique is generally slow and has low sensitivity, which limits its application in the clinical environment. Several challenges exist that limit the application of MRI in the clinical environment. Traditionally, the main limitations in MRI have been due to the manner in which data are sampled in clinical scans. The techniques of tiling have been applied with constant memory, loop invariant code motion, storing variables in registers, and using single-precision floating-point computations on the GPU kernels. The parallel structure of the reconstruction algorithms makes it suitable for parallel programming on GPUs. Accelerating this kind of algorithm can allow for more accurate image reconstruction while keeping computation times short enough for clinical use. Thus, the use of GPUs will enable improved trade-offs between data acquisition time, signal-to-noise ratio, and the severity of artifacts owing to nonideal physical effects during the MRI imaging experiment. © 2011

UR - http://www.scopus.com/inward/record.url?scp=84884437018&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84884437018&partnerID=8YFLogxK

U2 - 10.1016/B978-0-12-384988-5.00044-9

DO - 10.1016/B978-0-12-384988-5.00044-9

M3 - Chapter

AN - SCOPUS:84884437018

SN - 9780123849885

SP - 709

EP - 722

BT - GPU Computing Gems Emerald Edition

PB - Elsevier Inc.

ER -