Combined algorithmic and GPU acceleration for ultra-fast circular conebeam backprojection

Jeffrey Brokish, Paul Sack, Yoram Bresler

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

Abstract

In this paper, we describe the first implementation and performance of a fast O(N3logN) hierarchical backprojection algorithm for cone beam CT with a circular trajectory1,developed on a modern Graphics Processing Unit (GPU). The resulting tomographic backprojection system for 3D cone beam geometry combines speedup through algorithmic improvements provided by the hierarchical backprojection algorithm with speedup from a massively parallel hardware accelerator. For data parameters typical in diagnostic CT and using a mid-range GPU card, we report reconstruction speeds of up to 360 frames per second, and relative speedup of almost 6× compared to conventional backprojection on the same hardware. The significance of these results is twofold. First, they demonstrate that the reduction in operation counts demonstrated previously for the FHBP algorithm can be translated to a comparable run-time improvement in a massively parallel hardware implementation, while preserving stringent diagnostic image quality. Second, the dramatic speedup and throughput numbers achieved indicate the feasibility of systems based on this technology, which achieve real-time 3D reconstruction for state-of-the art diagnostic CT scanners with small footprint, high-reliability, and affordable cost.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2010
Subtitle of host publicationPhysics of Medical Imaging
EditionPART 3
DOIs
StatePublished - Dec 1 2010
EventMedical Imaging 2010: Physics of Medical Imaging - San Diego, CA, United States
Duration: Feb 15 2010Feb 18 2010

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
NumberPART 3
Volume7622
ISSN (Print)1605-7422

Other

OtherMedical Imaging 2010: Physics of Medical Imaging
CountryUnited States
CitySan Diego, CA
Period2/15/102/18/10

Fingerprint

hardware
Hardware
Cones
cones
Cone-Beam Computed Tomography
cards
footprints
preserving
Image quality
scanners
Particle accelerators
accelerators
Throughput
Technology
costs
Costs and Cost Analysis
Geometry
geometry
Graphics processing unit
Costs

Keywords

  • Circular cone-beam
  • Fast backprojection
  • Hierarchical algorithm
  • Tomography

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Brokish, J., Sack, P., & Bresler, Y. (2010). Combined algorithmic and GPU acceleration for ultra-fast circular conebeam backprojection. In Medical Imaging 2010: Physics of Medical Imaging (PART 3 ed.). [762256] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 7622, No. PART 3). https://doi.org/10.1117/12.844028

Combined algorithmic and GPU acceleration for ultra-fast circular conebeam backprojection. / Brokish, Jeffrey; Sack, Paul; Bresler, Yoram.

Medical Imaging 2010: Physics of Medical Imaging. PART 3. ed. 2010. 762256 (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 7622, No. PART 3).

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

Brokish, J, Sack, P & Bresler, Y 2010, Combined algorithmic and GPU acceleration for ultra-fast circular conebeam backprojection. in Medical Imaging 2010: Physics of Medical Imaging. PART 3 edn, 762256, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, no. PART 3, vol. 7622, Medical Imaging 2010: Physics of Medical Imaging, San Diego, CA, United States, 2/15/10. https://doi.org/10.1117/12.844028
Brokish J, Sack P, Bresler Y. Combined algorithmic and GPU acceleration for ultra-fast circular conebeam backprojection. In Medical Imaging 2010: Physics of Medical Imaging. PART 3 ed. 2010. 762256. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; PART 3). https://doi.org/10.1117/12.844028
Brokish, Jeffrey ; Sack, Paul ; Bresler, Yoram. / Combined algorithmic and GPU acceleration for ultra-fast circular conebeam backprojection. Medical Imaging 2010: Physics of Medical Imaging. PART 3. ed. 2010. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; PART 3).
@inproceedings{7de6926c90734ebf88223193abf2a48f,
title = "Combined algorithmic and GPU acceleration for ultra-fast circular conebeam backprojection",
abstract = "In this paper, we describe the first implementation and performance of a fast O(N3logN) hierarchical backprojection algorithm for cone beam CT with a circular trajectory1,developed on a modern Graphics Processing Unit (GPU). The resulting tomographic backprojection system for 3D cone beam geometry combines speedup through algorithmic improvements provided by the hierarchical backprojection algorithm with speedup from a massively parallel hardware accelerator. For data parameters typical in diagnostic CT and using a mid-range GPU card, we report reconstruction speeds of up to 360 frames per second, and relative speedup of almost 6× compared to conventional backprojection on the same hardware. The significance of these results is twofold. First, they demonstrate that the reduction in operation counts demonstrated previously for the FHBP algorithm can be translated to a comparable run-time improvement in a massively parallel hardware implementation, while preserving stringent diagnostic image quality. Second, the dramatic speedup and throughput numbers achieved indicate the feasibility of systems based on this technology, which achieve real-time 3D reconstruction for state-of-the art diagnostic CT scanners with small footprint, high-reliability, and affordable cost.",
keywords = "Circular cone-beam, Fast backprojection, Hierarchical algorithm, Tomography",
author = "Jeffrey Brokish and Paul Sack and Yoram Bresler",
year = "2010",
month = "12",
day = "1",
doi = "10.1117/12.844028",
language = "English (US)",
isbn = "9780819480231",
series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
number = "PART 3",
booktitle = "Medical Imaging 2010",
edition = "PART 3",

}

TY - GEN

T1 - Combined algorithmic and GPU acceleration for ultra-fast circular conebeam backprojection

AU - Brokish, Jeffrey

AU - Sack, Paul

AU - Bresler, Yoram

PY - 2010/12/1

Y1 - 2010/12/1

N2 - In this paper, we describe the first implementation and performance of a fast O(N3logN) hierarchical backprojection algorithm for cone beam CT with a circular trajectory1,developed on a modern Graphics Processing Unit (GPU). The resulting tomographic backprojection system for 3D cone beam geometry combines speedup through algorithmic improvements provided by the hierarchical backprojection algorithm with speedup from a massively parallel hardware accelerator. For data parameters typical in diagnostic CT and using a mid-range GPU card, we report reconstruction speeds of up to 360 frames per second, and relative speedup of almost 6× compared to conventional backprojection on the same hardware. The significance of these results is twofold. First, they demonstrate that the reduction in operation counts demonstrated previously for the FHBP algorithm can be translated to a comparable run-time improvement in a massively parallel hardware implementation, while preserving stringent diagnostic image quality. Second, the dramatic speedup and throughput numbers achieved indicate the feasibility of systems based on this technology, which achieve real-time 3D reconstruction for state-of-the art diagnostic CT scanners with small footprint, high-reliability, and affordable cost.

AB - In this paper, we describe the first implementation and performance of a fast O(N3logN) hierarchical backprojection algorithm for cone beam CT with a circular trajectory1,developed on a modern Graphics Processing Unit (GPU). The resulting tomographic backprojection system for 3D cone beam geometry combines speedup through algorithmic improvements provided by the hierarchical backprojection algorithm with speedup from a massively parallel hardware accelerator. For data parameters typical in diagnostic CT and using a mid-range GPU card, we report reconstruction speeds of up to 360 frames per second, and relative speedup of almost 6× compared to conventional backprojection on the same hardware. The significance of these results is twofold. First, they demonstrate that the reduction in operation counts demonstrated previously for the FHBP algorithm can be translated to a comparable run-time improvement in a massively parallel hardware implementation, while preserving stringent diagnostic image quality. Second, the dramatic speedup and throughput numbers achieved indicate the feasibility of systems based on this technology, which achieve real-time 3D reconstruction for state-of-the art diagnostic CT scanners with small footprint, high-reliability, and affordable cost.

KW - Circular cone-beam

KW - Fast backprojection

KW - Hierarchical algorithm

KW - Tomography

UR - http://www.scopus.com/inward/record.url?scp=84873899341&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84873899341&partnerID=8YFLogxK

U2 - 10.1117/12.844028

DO - 10.1117/12.844028

M3 - Conference contribution

AN - SCOPUS:84873899341

SN - 9780819480231

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

BT - Medical Imaging 2010

ER -