GPU-based fast low-dose cone beam CT reconstruction via total variation

Xun Jia, Yifei Lou, John Lewis, Ruijiang Li, Xuejun Gu, Chunhua Men, William Y. Song, Steve B. Jiang

Research output: Contribution to journalArticle

50 Citations (Scopus)

Abstract

X-ray imaging dose from serial Cone-beam CT (CBCT) scans raises a clinical concern in most image guided radiation therapy procedures. The goal of this paper is to develop a fast GPU-based algorithm to reconstruct high quality CBCT images from undersampled and noisy projection data so as to lower the imaging dose. The CBCT is reconstructed by minimizing an energy functional consisting of a data fidelity term and a total variation regularization term. We develop a GPU-friendly version of a forward-backward splitting algorithm to solve this problem. A multi-grid technique is also employed. We test our CBCT reconstruction algorithm on a digital phantom and a head-and-neck patient case. The performance under low mAs is also validated using physical phantoms. It is found that 40 x-ray projections are sufficient to reconstruct CBCT images with satisfactory quality for clinical purposes. Phantom experiments indicate that CBCT images can be successfully reconstructed under 0.1 mAs/projection. Comparing with the widely used head-and-neck scanning protocol of about 360 projections with 0.4 mAs/projection, an overall 36 times dose reduction has been achieved. The reconstruction time is about 130 sec on an NVIDIA Tesla C1060 GPU card, which is estimated ∼ 100 times faster than similar regularized iterative reconstruction approaches.

Original languageEnglish (US)
Pages (from-to)139-154
Number of pages16
JournalJournal of X-Ray Science and Technology
Volume19
Issue number2
DOIs
StatePublished - 2011

Fingerprint

Cone-Beam Computed Tomography
Dosimetry
Cones
cones
projection
dosage
Neck
Head
X-Rays
Image-Guided Radiotherapy
Imaging techniques
X rays
Radiotherapy
cards
Graphics processing unit
radiation therapy
x rays
grids
Scanning
scanning

Keywords

  • Cone beam CT
  • GPU
  • total variation

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Condensed Matter Physics
  • Radiation
  • Electrical and Electronic Engineering
  • Instrumentation

Cite this

GPU-based fast low-dose cone beam CT reconstruction via total variation. / Jia, Xun; Lou, Yifei; Lewis, John; Li, Ruijiang; Gu, Xuejun; Men, Chunhua; Song, William Y.; Jiang, Steve B.

In: Journal of X-Ray Science and Technology, Vol. 19, No. 2, 2011, p. 139-154.

Research output: Contribution to journalArticle

Jia, Xun ; Lou, Yifei ; Lewis, John ; Li, Ruijiang ; Gu, Xuejun ; Men, Chunhua ; Song, William Y. ; Jiang, Steve B. / GPU-based fast low-dose cone beam CT reconstruction via total variation. In: Journal of X-Ray Science and Technology. 2011 ; Vol. 19, No. 2. pp. 139-154.
@article{049a71b2f7fa4fbea2b4ba0901c9246f,
title = "GPU-based fast low-dose cone beam CT reconstruction via total variation",
abstract = "X-ray imaging dose from serial Cone-beam CT (CBCT) scans raises a clinical concern in most image guided radiation therapy procedures. The goal of this paper is to develop a fast GPU-based algorithm to reconstruct high quality CBCT images from undersampled and noisy projection data so as to lower the imaging dose. The CBCT is reconstructed by minimizing an energy functional consisting of a data fidelity term and a total variation regularization term. We develop a GPU-friendly version of a forward-backward splitting algorithm to solve this problem. A multi-grid technique is also employed. We test our CBCT reconstruction algorithm on a digital phantom and a head-and-neck patient case. The performance under low mAs is also validated using physical phantoms. It is found that 40 x-ray projections are sufficient to reconstruct CBCT images with satisfactory quality for clinical purposes. Phantom experiments indicate that CBCT images can be successfully reconstructed under 0.1 mAs/projection. Comparing with the widely used head-and-neck scanning protocol of about 360 projections with 0.4 mAs/projection, an overall 36 times dose reduction has been achieved. The reconstruction time is about 130 sec on an NVIDIA Tesla C1060 GPU card, which is estimated ∼ 100 times faster than similar regularized iterative reconstruction approaches.",
keywords = "Cone beam CT, GPU, total variation",
author = "Xun Jia and Yifei Lou and John Lewis and Ruijiang Li and Xuejun Gu and Chunhua Men and Song, {William Y.} and Jiang, {Steve B.}",
year = "2011",
doi = "10.3233/XST-2011-0283",
language = "English (US)",
volume = "19",
pages = "139--154",
journal = "Journal of X-Ray Science and Technology",
issn = "0895-3996",
publisher = "IOS Press",
number = "2",

}

TY - JOUR

T1 - GPU-based fast low-dose cone beam CT reconstruction via total variation

AU - Jia, Xun

AU - Lou, Yifei

AU - Lewis, John

AU - Li, Ruijiang

AU - Gu, Xuejun

AU - Men, Chunhua

AU - Song, William Y.

AU - Jiang, Steve B.

PY - 2011

Y1 - 2011

N2 - X-ray imaging dose from serial Cone-beam CT (CBCT) scans raises a clinical concern in most image guided radiation therapy procedures. The goal of this paper is to develop a fast GPU-based algorithm to reconstruct high quality CBCT images from undersampled and noisy projection data so as to lower the imaging dose. The CBCT is reconstructed by minimizing an energy functional consisting of a data fidelity term and a total variation regularization term. We develop a GPU-friendly version of a forward-backward splitting algorithm to solve this problem. A multi-grid technique is also employed. We test our CBCT reconstruction algorithm on a digital phantom and a head-and-neck patient case. The performance under low mAs is also validated using physical phantoms. It is found that 40 x-ray projections are sufficient to reconstruct CBCT images with satisfactory quality for clinical purposes. Phantom experiments indicate that CBCT images can be successfully reconstructed under 0.1 mAs/projection. Comparing with the widely used head-and-neck scanning protocol of about 360 projections with 0.4 mAs/projection, an overall 36 times dose reduction has been achieved. The reconstruction time is about 130 sec on an NVIDIA Tesla C1060 GPU card, which is estimated ∼ 100 times faster than similar regularized iterative reconstruction approaches.

AB - X-ray imaging dose from serial Cone-beam CT (CBCT) scans raises a clinical concern in most image guided radiation therapy procedures. The goal of this paper is to develop a fast GPU-based algorithm to reconstruct high quality CBCT images from undersampled and noisy projection data so as to lower the imaging dose. The CBCT is reconstructed by minimizing an energy functional consisting of a data fidelity term and a total variation regularization term. We develop a GPU-friendly version of a forward-backward splitting algorithm to solve this problem. A multi-grid technique is also employed. We test our CBCT reconstruction algorithm on a digital phantom and a head-and-neck patient case. The performance under low mAs is also validated using physical phantoms. It is found that 40 x-ray projections are sufficient to reconstruct CBCT images with satisfactory quality for clinical purposes. Phantom experiments indicate that CBCT images can be successfully reconstructed under 0.1 mAs/projection. Comparing with the widely used head-and-neck scanning protocol of about 360 projections with 0.4 mAs/projection, an overall 36 times dose reduction has been achieved. The reconstruction time is about 130 sec on an NVIDIA Tesla C1060 GPU card, which is estimated ∼ 100 times faster than similar regularized iterative reconstruction approaches.

KW - Cone beam CT

KW - GPU

KW - total variation

UR - http://www.scopus.com/inward/record.url?scp=79959269966&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79959269966&partnerID=8YFLogxK

U2 - 10.3233/XST-2011-0283

DO - 10.3233/XST-2011-0283

M3 - Article

C2 - 21606579

AN - SCOPUS:79959269966

VL - 19

SP - 139

EP - 154

JO - Journal of X-Ray Science and Technology

JF - Journal of X-Ray Science and Technology

SN - 0895-3996

IS - 2

ER -