An Inter-Projection Interpolation (IPI) approach with geometric model restriction to reduce image dose in Cone Beam CT (CBCT)

Hong Zhang, Fengchong Kong, Lei Ren, JianYue Jin

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

3 Citations (Scopus)

Abstract

Cone beam computed tomography (CBCT) imaging is a key step in image guided radiation therapy (IGRT) to improve tumor targeting. The quality and imaging dose of CBCT are two important factors. However, X-ray scatter in the large cone beam field usually induces image artifacts and degrades the image quality for CBCT. A synchronized moving grid (SMOG) approach has recently been proposed to resolve this issue and shows great promise. However, the SMOG technique requires two projections in the same gantry angle to obtain full information due to signal blockage by the grid. This study aims to develop an inter-projection interpolation (IPI) method to estimate the blocked image information. This approach will require only one projection in each gantry angle, thus reducing the scan time and patient dose. IPI is also potentially suitable for sparse-view CBCT reconstruction to reduce the imaging dose. To be compared with other state-of-the-art spatial interpolation (called inpainting) methods in terms of signal-to-noise ratio (SNR) on a Catphan and head phantoms, IPI increases SNR from 15.3dB and 12.7dB to 29.0dB and 28.1dB, respectively. The SNR of IPI on sparse-view CBCT reconstruction can achieve from 28dB to 17dB for undersample projection sets with gantry angle interval varying from 1 to 3 degrees for both phantoms.

Original languageEnglish (US)
Title of host publicationComputational Modeling of Objects Presented in Images
Subtitle of host publicationFundamentals, Methods, and Applications - 4th International Conference, CompIMAGE 2014, Proceedings
PublisherSpringer Verlag
Pages12-23
Number of pages12
ISBN (Print)9783319099934
DOIs
StatePublished - Jan 1 2014
Event4th International Conference on Computational Modeling of Objects Presented in Images: Fundamentals, Methods, and Applications, CompIMAGE 2014 - Pittsburgh, PA, United States
Duration: Sep 3 2014Sep 5 2014

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8641 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other4th International Conference on Computational Modeling of Objects Presented in Images: Fundamentals, Methods, and Applications, CompIMAGE 2014
CountryUnited States
CityPittsburgh, PA
Period9/3/149/5/14

Fingerprint

Geometric Model
Cones
Dose
Interpolation
Cone
Interpolate
Projection
Restriction
Moving Grid
Signal to noise ratio
Imaging
Phantom
Imaging techniques
Angle
Spatial Interpolation
Radiation Therapy
Inpainting
Radiotherapy
Computed Tomography
Interpolation Method

Keywords

  • CBCT
  • Dose reduction
  • Geometric model
  • Interpolation
  • Moving grids
  • SMOG
  • Scatter correction

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Zhang, H., Kong, F., Ren, L., & Jin, J. (2014). An Inter-Projection Interpolation (IPI) approach with geometric model restriction to reduce image dose in Cone Beam CT (CBCT). In Computational Modeling of Objects Presented in Images: Fundamentals, Methods, and Applications - 4th International Conference, CompIMAGE 2014, Proceedings (pp. 12-23). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8641 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-09994-1_2

An Inter-Projection Interpolation (IPI) approach with geometric model restriction to reduce image dose in Cone Beam CT (CBCT). / Zhang, Hong; Kong, Fengchong; Ren, Lei; Jin, JianYue.

Computational Modeling of Objects Presented in Images: Fundamentals, Methods, and Applications - 4th International Conference, CompIMAGE 2014, Proceedings. Springer Verlag, 2014. p. 12-23 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8641 LNCS).

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

Zhang, H, Kong, F, Ren, L & Jin, J 2014, An Inter-Projection Interpolation (IPI) approach with geometric model restriction to reduce image dose in Cone Beam CT (CBCT). in Computational Modeling of Objects Presented in Images: Fundamentals, Methods, and Applications - 4th International Conference, CompIMAGE 2014, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8641 LNCS, Springer Verlag, pp. 12-23, 4th International Conference on Computational Modeling of Objects Presented in Images: Fundamentals, Methods, and Applications, CompIMAGE 2014, Pittsburgh, PA, United States, 9/3/14. https://doi.org/10.1007/978-3-319-09994-1_2
Zhang H, Kong F, Ren L, Jin J. An Inter-Projection Interpolation (IPI) approach with geometric model restriction to reduce image dose in Cone Beam CT (CBCT). In Computational Modeling of Objects Presented in Images: Fundamentals, Methods, and Applications - 4th International Conference, CompIMAGE 2014, Proceedings. Springer Verlag. 2014. p. 12-23. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-09994-1_2
Zhang, Hong ; Kong, Fengchong ; Ren, Lei ; Jin, JianYue. / An Inter-Projection Interpolation (IPI) approach with geometric model restriction to reduce image dose in Cone Beam CT (CBCT). Computational Modeling of Objects Presented in Images: Fundamentals, Methods, and Applications - 4th International Conference, CompIMAGE 2014, Proceedings. Springer Verlag, 2014. pp. 12-23 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{e79efb81bb3241898652d6329c66b023,
title = "An Inter-Projection Interpolation (IPI) approach with geometric model restriction to reduce image dose in Cone Beam CT (CBCT)",
abstract = "Cone beam computed tomography (CBCT) imaging is a key step in image guided radiation therapy (IGRT) to improve tumor targeting. The quality and imaging dose of CBCT are two important factors. However, X-ray scatter in the large cone beam field usually induces image artifacts and degrades the image quality for CBCT. A synchronized moving grid (SMOG) approach has recently been proposed to resolve this issue and shows great promise. However, the SMOG technique requires two projections in the same gantry angle to obtain full information due to signal blockage by the grid. This study aims to develop an inter-projection interpolation (IPI) method to estimate the blocked image information. This approach will require only one projection in each gantry angle, thus reducing the scan time and patient dose. IPI is also potentially suitable for sparse-view CBCT reconstruction to reduce the imaging dose. To be compared with other state-of-the-art spatial interpolation (called inpainting) methods in terms of signal-to-noise ratio (SNR) on a Catphan and head phantoms, IPI increases SNR from 15.3dB and 12.7dB to 29.0dB and 28.1dB, respectively. The SNR of IPI on sparse-view CBCT reconstruction can achieve from 28dB to 17dB for undersample projection sets with gantry angle interval varying from 1 to 3 degrees for both phantoms.",
keywords = "CBCT, Dose reduction, Geometric model, Interpolation, Moving grids, SMOG, Scatter correction",
author = "Hong Zhang and Fengchong Kong and Lei Ren and JianYue Jin",
year = "2014",
month = "1",
day = "1",
doi = "10.1007/978-3-319-09994-1_2",
language = "English (US)",
isbn = "9783319099934",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "12--23",
booktitle = "Computational Modeling of Objects Presented in Images",

}

TY - GEN

T1 - An Inter-Projection Interpolation (IPI) approach with geometric model restriction to reduce image dose in Cone Beam CT (CBCT)

AU - Zhang, Hong

AU - Kong, Fengchong

AU - Ren, Lei

AU - Jin, JianYue

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Cone beam computed tomography (CBCT) imaging is a key step in image guided radiation therapy (IGRT) to improve tumor targeting. The quality and imaging dose of CBCT are two important factors. However, X-ray scatter in the large cone beam field usually induces image artifacts and degrades the image quality for CBCT. A synchronized moving grid (SMOG) approach has recently been proposed to resolve this issue and shows great promise. However, the SMOG technique requires two projections in the same gantry angle to obtain full information due to signal blockage by the grid. This study aims to develop an inter-projection interpolation (IPI) method to estimate the blocked image information. This approach will require only one projection in each gantry angle, thus reducing the scan time and patient dose. IPI is also potentially suitable for sparse-view CBCT reconstruction to reduce the imaging dose. To be compared with other state-of-the-art spatial interpolation (called inpainting) methods in terms of signal-to-noise ratio (SNR) on a Catphan and head phantoms, IPI increases SNR from 15.3dB and 12.7dB to 29.0dB and 28.1dB, respectively. The SNR of IPI on sparse-view CBCT reconstruction can achieve from 28dB to 17dB for undersample projection sets with gantry angle interval varying from 1 to 3 degrees for both phantoms.

AB - Cone beam computed tomography (CBCT) imaging is a key step in image guided radiation therapy (IGRT) to improve tumor targeting. The quality and imaging dose of CBCT are two important factors. However, X-ray scatter in the large cone beam field usually induces image artifacts and degrades the image quality for CBCT. A synchronized moving grid (SMOG) approach has recently been proposed to resolve this issue and shows great promise. However, the SMOG technique requires two projections in the same gantry angle to obtain full information due to signal blockage by the grid. This study aims to develop an inter-projection interpolation (IPI) method to estimate the blocked image information. This approach will require only one projection in each gantry angle, thus reducing the scan time and patient dose. IPI is also potentially suitable for sparse-view CBCT reconstruction to reduce the imaging dose. To be compared with other state-of-the-art spatial interpolation (called inpainting) methods in terms of signal-to-noise ratio (SNR) on a Catphan and head phantoms, IPI increases SNR from 15.3dB and 12.7dB to 29.0dB and 28.1dB, respectively. The SNR of IPI on sparse-view CBCT reconstruction can achieve from 28dB to 17dB for undersample projection sets with gantry angle interval varying from 1 to 3 degrees for both phantoms.

KW - CBCT

KW - Dose reduction

KW - Geometric model

KW - Interpolation

KW - Moving grids

KW - SMOG

KW - Scatter correction

UR - http://www.scopus.com/inward/record.url?scp=84905827194&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84905827194&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-09994-1_2

DO - 10.1007/978-3-319-09994-1_2

M3 - Conference contribution

SN - 9783319099934

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 12

EP - 23

BT - Computational Modeling of Objects Presented in Images

PB - Springer Verlag

ER -