GPU-based motion correction of contrast-enhanced liver MRI scans

An OpenCL implementation

Jihun Oh, Diego R Martin, Oskar Škrinjar

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

3 Citations (Scopus)

Abstract

Clinical diagnosis and quantification of liver disease have been improved through the development of techniques using contrast-enhanced liver MRI sequences. To qualitatively or quantitatively analyze such image sequences, one first needs to correct for rigid and non-rigid motion of the liver. For motion correction of the liver, we have employed bi-directional local correlation coefficient Demons, which is a variation of the original Demons method. However, despite the intrinsic speed of the Demons method, the run-time on the order of an hour of its CPU-based implementation is not sufficiently short for a regular clinical use. For this reason we implemented the method on a graphics processing unit (GPU) using OpenCL. On NVIDIA GTX 260M, which is a laptop GPU, we achieved sub-minute runtime for the motion correction of typical liver MRI scans, which was 50 times faster than its CPU-based implementation. A sub-minute runtime of liver MRI motion correction allows for its regular clinical use.

Original languageEnglish (US)
Title of host publicationProceedings - International Symposium on Biomedical Imaging
Pages783-786
Number of pages4
DOIs
StatePublished - 2011
Externally publishedYes
Event2011 8th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI'11 - Chicago, IL, United States
Duration: Mar 30 2011Apr 2 2011

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

Liver
Magnetic Resonance Imaging
Magnetic resonance imaging
Program processors
Liver Diseases
Graphics processing unit

Keywords

  • contrast-enhanced liver MRI
  • Demons
  • GPU
  • graphics processing unit
  • image registration
  • OpenCL

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Oh, J., Martin, D. R., & Škrinjar, O. (2011). GPU-based motion correction of contrast-enhanced liver MRI scans: An OpenCL implementation. In Proceedings - International Symposium on Biomedical Imaging (pp. 783-786). [5872522] https://doi.org/10.1109/ISBI.2011.5872522

GPU-based motion correction of contrast-enhanced liver MRI scans : An OpenCL implementation. / Oh, Jihun; Martin, Diego R; Škrinjar, Oskar.

Proceedings - International Symposium on Biomedical Imaging. 2011. p. 783-786 5872522.

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

Oh, J, Martin, DR & Škrinjar, O 2011, GPU-based motion correction of contrast-enhanced liver MRI scans: An OpenCL implementation. in Proceedings - International Symposium on Biomedical Imaging., 5872522, pp. 783-786, 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.2011.5872522
Oh J, Martin DR, Škrinjar O. GPU-based motion correction of contrast-enhanced liver MRI scans: An OpenCL implementation. In Proceedings - International Symposium on Biomedical Imaging. 2011. p. 783-786. 5872522 https://doi.org/10.1109/ISBI.2011.5872522
Oh, Jihun ; Martin, Diego R ; Škrinjar, Oskar. / GPU-based motion correction of contrast-enhanced liver MRI scans : An OpenCL implementation. Proceedings - International Symposium on Biomedical Imaging. 2011. pp. 783-786
@inproceedings{56fc6c8f125f41848e07427d546e03c6,
title = "GPU-based motion correction of contrast-enhanced liver MRI scans: An OpenCL implementation",
abstract = "Clinical diagnosis and quantification of liver disease have been improved through the development of techniques using contrast-enhanced liver MRI sequences. To qualitatively or quantitatively analyze such image sequences, one first needs to correct for rigid and non-rigid motion of the liver. For motion correction of the liver, we have employed bi-directional local correlation coefficient Demons, which is a variation of the original Demons method. However, despite the intrinsic speed of the Demons method, the run-time on the order of an hour of its CPU-based implementation is not sufficiently short for a regular clinical use. For this reason we implemented the method on a graphics processing unit (GPU) using OpenCL. On NVIDIA GTX 260M, which is a laptop GPU, we achieved sub-minute runtime for the motion correction of typical liver MRI scans, which was 50 times faster than its CPU-based implementation. A sub-minute runtime of liver MRI motion correction allows for its regular clinical use.",
keywords = "contrast-enhanced liver MRI, Demons, GPU, graphics processing unit, image registration, OpenCL",
author = "Jihun Oh and Martin, {Diego R} and Oskar Škrinjar",
year = "2011",
doi = "10.1109/ISBI.2011.5872522",
language = "English (US)",
isbn = "9781424441280",
pages = "783--786",
booktitle = "Proceedings - International Symposium on Biomedical Imaging",

}

TY - GEN

T1 - GPU-based motion correction of contrast-enhanced liver MRI scans

T2 - An OpenCL implementation

AU - Oh, Jihun

AU - Martin, Diego R

AU - Škrinjar, Oskar

PY - 2011

Y1 - 2011

N2 - Clinical diagnosis and quantification of liver disease have been improved through the development of techniques using contrast-enhanced liver MRI sequences. To qualitatively or quantitatively analyze such image sequences, one first needs to correct for rigid and non-rigid motion of the liver. For motion correction of the liver, we have employed bi-directional local correlation coefficient Demons, which is a variation of the original Demons method. However, despite the intrinsic speed of the Demons method, the run-time on the order of an hour of its CPU-based implementation is not sufficiently short for a regular clinical use. For this reason we implemented the method on a graphics processing unit (GPU) using OpenCL. On NVIDIA GTX 260M, which is a laptop GPU, we achieved sub-minute runtime for the motion correction of typical liver MRI scans, which was 50 times faster than its CPU-based implementation. A sub-minute runtime of liver MRI motion correction allows for its regular clinical use.

AB - Clinical diagnosis and quantification of liver disease have been improved through the development of techniques using contrast-enhanced liver MRI sequences. To qualitatively or quantitatively analyze such image sequences, one first needs to correct for rigid and non-rigid motion of the liver. For motion correction of the liver, we have employed bi-directional local correlation coefficient Demons, which is a variation of the original Demons method. However, despite the intrinsic speed of the Demons method, the run-time on the order of an hour of its CPU-based implementation is not sufficiently short for a regular clinical use. For this reason we implemented the method on a graphics processing unit (GPU) using OpenCL. On NVIDIA GTX 260M, which is a laptop GPU, we achieved sub-minute runtime for the motion correction of typical liver MRI scans, which was 50 times faster than its CPU-based implementation. A sub-minute runtime of liver MRI motion correction allows for its regular clinical use.

KW - contrast-enhanced liver MRI

KW - Demons

KW - GPU

KW - graphics processing unit

KW - image registration

KW - OpenCL

UR - http://www.scopus.com/inward/record.url?scp=80055045994&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80055045994&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2011.5872522

DO - 10.1109/ISBI.2011.5872522

M3 - Conference contribution

SN - 9781424441280

SP - 783

EP - 786

BT - Proceedings - International Symposium on Biomedical Imaging

ER -