Implementation and evaluation of various demons deformable image registration algorithms on a GPU

Xuejun Gu, Hubert Pan, Yun Liang, Richard Castillo, Deshan Yang, Dongju Choi, Edward Castillo, Amitava Majumdar, Thomas Guerrero, Steve B. Jiang

Research output: Contribution to journalArticle

170 Citations (Scopus)

Abstract

Online adaptive radiation therapy (ART) promises the ability to deliver an optimal treatment in response to daily patient anatomic variation. A major technical barrier for the clinical implementation of online ART is the requirement of rapid image segmentation. Deformable image registration (DIR) has been used as an automated segmentation method to transfer tumor/organ contours from the planning image to daily images. However, the current computational time of DIR is insufficient for online ART. In this work, this issue is addressed by using computer graphics processing units (GPUs). A gray-scale-based DIR algorithm called demons and five of its variants were implemented on GPUs using the compute unified device architecture (CUDA) programming environment. The spatial accuracy of these algorithms was evaluated over five sets of pulmonary 4D CT images with an average size of 256 × 256 × 100 and more than 1100 expert-determined landmark point pairs each. For all the testing scenarios presented in this paper, the GPU-based DIR computation required around 7 to 11 s to yield an average 3D error ranging from 1.5 to 1.8 mm. It is interesting to find out that the original passive force demons algorithms outperform subsequently proposed variants based on the combination of accuracy, efficiency and ease of implementation.

Original languageEnglish (US)
Pages (from-to)207-219
Number of pages13
JournalPhysics in Medicine and Biology
Volume55
Issue number1
DOIs
StatePublished - Jan 7 2010
Externally publishedYes

Fingerprint

Radiotherapy
Four-Dimensional Computed Tomography
Computer Graphics
Anatomic Variation
Aptitude
Efficiency
Equipment and Supplies
Lung
Neoplasms
Therapeutics

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology

Cite this

Implementation and evaluation of various demons deformable image registration algorithms on a GPU. / Gu, Xuejun; Pan, Hubert; Liang, Yun; Castillo, Richard; Yang, Deshan; Choi, Dongju; Castillo, Edward; Majumdar, Amitava; Guerrero, Thomas; Jiang, Steve B.

In: Physics in Medicine and Biology, Vol. 55, No. 1, 07.01.2010, p. 207-219.

Research output: Contribution to journalArticle

Gu, X, Pan, H, Liang, Y, Castillo, R, Yang, D, Choi, D, Castillo, E, Majumdar, A, Guerrero, T & Jiang, SB 2010, 'Implementation and evaluation of various demons deformable image registration algorithms on a GPU', Physics in Medicine and Biology, vol. 55, no. 1, pp. 207-219. https://doi.org/10.1088/0031-9155/55/1/012
Gu, Xuejun ; Pan, Hubert ; Liang, Yun ; Castillo, Richard ; Yang, Deshan ; Choi, Dongju ; Castillo, Edward ; Majumdar, Amitava ; Guerrero, Thomas ; Jiang, Steve B. / Implementation and evaluation of various demons deformable image registration algorithms on a GPU. In: Physics in Medicine and Biology. 2010 ; Vol. 55, No. 1. pp. 207-219.
@article{0f18ec710d16444590b7271f56eac04e,
title = "Implementation and evaluation of various demons deformable image registration algorithms on a GPU",
abstract = "Online adaptive radiation therapy (ART) promises the ability to deliver an optimal treatment in response to daily patient anatomic variation. A major technical barrier for the clinical implementation of online ART is the requirement of rapid image segmentation. Deformable image registration (DIR) has been used as an automated segmentation method to transfer tumor/organ contours from the planning image to daily images. However, the current computational time of DIR is insufficient for online ART. In this work, this issue is addressed by using computer graphics processing units (GPUs). A gray-scale-based DIR algorithm called demons and five of its variants were implemented on GPUs using the compute unified device architecture (CUDA) programming environment. The spatial accuracy of these algorithms was evaluated over five sets of pulmonary 4D CT images with an average size of 256 × 256 × 100 and more than 1100 expert-determined landmark point pairs each. For all the testing scenarios presented in this paper, the GPU-based DIR computation required around 7 to 11 s to yield an average 3D error ranging from 1.5 to 1.8 mm. It is interesting to find out that the original passive force demons algorithms outperform subsequently proposed variants based on the combination of accuracy, efficiency and ease of implementation.",
author = "Xuejun Gu and Hubert Pan and Yun Liang and Richard Castillo and Deshan Yang and Dongju Choi and Edward Castillo and Amitava Majumdar and Thomas Guerrero and Jiang, {Steve B.}",
year = "2010",
month = "1",
day = "7",
doi = "10.1088/0031-9155/55/1/012",
language = "English (US)",
volume = "55",
pages = "207--219",
journal = "Physics in Medicine and Biology",
issn = "0031-9155",
publisher = "IOP Publishing Ltd.",
number = "1",

}

TY - JOUR

T1 - Implementation and evaluation of various demons deformable image registration algorithms on a GPU

AU - Gu, Xuejun

AU - Pan, Hubert

AU - Liang, Yun

AU - Castillo, Richard

AU - Yang, Deshan

AU - Choi, Dongju

AU - Castillo, Edward

AU - Majumdar, Amitava

AU - Guerrero, Thomas

AU - Jiang, Steve B.

PY - 2010/1/7

Y1 - 2010/1/7

N2 - Online adaptive radiation therapy (ART) promises the ability to deliver an optimal treatment in response to daily patient anatomic variation. A major technical barrier for the clinical implementation of online ART is the requirement of rapid image segmentation. Deformable image registration (DIR) has been used as an automated segmentation method to transfer tumor/organ contours from the planning image to daily images. However, the current computational time of DIR is insufficient for online ART. In this work, this issue is addressed by using computer graphics processing units (GPUs). A gray-scale-based DIR algorithm called demons and five of its variants were implemented on GPUs using the compute unified device architecture (CUDA) programming environment. The spatial accuracy of these algorithms was evaluated over five sets of pulmonary 4D CT images with an average size of 256 × 256 × 100 and more than 1100 expert-determined landmark point pairs each. For all the testing scenarios presented in this paper, the GPU-based DIR computation required around 7 to 11 s to yield an average 3D error ranging from 1.5 to 1.8 mm. It is interesting to find out that the original passive force demons algorithms outperform subsequently proposed variants based on the combination of accuracy, efficiency and ease of implementation.

AB - Online adaptive radiation therapy (ART) promises the ability to deliver an optimal treatment in response to daily patient anatomic variation. A major technical barrier for the clinical implementation of online ART is the requirement of rapid image segmentation. Deformable image registration (DIR) has been used as an automated segmentation method to transfer tumor/organ contours from the planning image to daily images. However, the current computational time of DIR is insufficient for online ART. In this work, this issue is addressed by using computer graphics processing units (GPUs). A gray-scale-based DIR algorithm called demons and five of its variants were implemented on GPUs using the compute unified device architecture (CUDA) programming environment. The spatial accuracy of these algorithms was evaluated over five sets of pulmonary 4D CT images with an average size of 256 × 256 × 100 and more than 1100 expert-determined landmark point pairs each. For all the testing scenarios presented in this paper, the GPU-based DIR computation required around 7 to 11 s to yield an average 3D error ranging from 1.5 to 1.8 mm. It is interesting to find out that the original passive force demons algorithms outperform subsequently proposed variants based on the combination of accuracy, efficiency and ease of implementation.

UR - http://www.scopus.com/inward/record.url?scp=77649198532&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77649198532&partnerID=8YFLogxK

U2 - 10.1088/0031-9155/55/1/012

DO - 10.1088/0031-9155/55/1/012

M3 - Article

VL - 55

SP - 207

EP - 219

JO - Physics in Medicine and Biology

JF - Physics in Medicine and Biology

SN - 0031-9155

IS - 1

ER -