Iterative reconstruction via preserved structures approach for CT images with limited scan angles

Anush Kolakalur; Branislav Vuksanovic

doi:10.1145/3332340.3332354

Iterative reconstruction via preserved structures approach for CT images with limited scan angles

Anush Kolakalur, Branislav Vuksanovic

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

430 Downloads (Pure)

Abstract

Iterative image reconstruction techniques for computer tomography (CT) are finite iterations of forward-projections and backward projections. One of the major concerns related to this method is deterioration of the reconstructed images due to various image structure deformations during this procedure. This is usually manifested by blotchy and pixelated appearances of the reconstructed image with the effects becoming more pronounced for low and ultra-low scan angles. This paper proposes a new approach for the reconstruction of CT images ensuring the preservation of structural details and reduced image deterioration and deformation. We call this method iterative reconstruction through preserved structures (IR-PS). The results achieved using proposed IR-PS method are evaluated via RMSE (Root Mean Square Error) measure and SSIM (Structural SIMiliarity) index suggesting improvement in quality of reconstructed images.

Original language	English
Title of host publication	IMIP '19: Proceedings of the 2019 International Conference on Intelligent Medicine and Image Processing
Publisher	Association for Computing Machinery
Pages	72-77
Number of pages	6
ISBN (Print)	978-1-4503-6269-6
DOIs	https://doi.org/10.1145/3332340.3332354
Publication status	Published - 22 Apr 2019
Event	3rd International Conference on Frontiers of Image Processing - Florence, Italy Duration: 16 Mar 2019 → 18 Mar 2019 http://www.icfip.org/

Conference

Conference	3rd International Conference on Frontiers of Image Processing
Abbreviated title	ICFIP 2019
Country/Territory	Italy
City	Florence
Period	16/03/19 → 18/03/19
Internet address	http://www.icfip.org/

Keywords

computed tomography
iterative image reconstruction
image degradation
limited scan angle
CT reconstruction

Access to Document

10.1145/3332340.3332354

P004paperBV
© Association for Computing Machinery 2019. This is the author's version of the work. It is posted here for your personal use. Not for redistribution. The definitive Version of Record was published in Proceedings of the 2019 International Conference on Intelligent Medicine and Image Processing, http://dx.doi.org/10.1145/3332340.3332354.
Accepted author manuscript (Post-print), 599 KB

Cite this

@inproceedings{54d375668f1445f0a7886dfc6c332789,

title = "Iterative reconstruction via preserved structures approach for CT images with limited scan angles",

abstract = "Iterative image reconstruction techniques for computer tomography (CT) are finite iterations of forward-projections and backward projections. One of the major concerns related to this method is deterioration of the reconstructed images due to various image structure deformations during this procedure. This is usually manifested by blotchy and pixelated appearances of the reconstructed image with the effects becoming more pronounced for low and ultra-low scan angles. This paper proposes a new approach for the reconstruction of CT images ensuring the preservation of structural details and reduced image deterioration and deformation. We call this method iterative reconstruction through preserved structures (IR-PS). The results achieved using proposed IR-PS method are evaluated via RMSE (Root Mean Square Error) measure and SSIM (Structural SIMiliarity) index suggesting improvement in quality of reconstructed images.",

keywords = "computed tomography, iterative image reconstruction, image degradation, limited scan angle, CT reconstruction",

author = "Anush Kolakalur and Branislav Vuksanovic",

year = "2019",

month = apr,

day = "22",

doi = "10.1145/3332340.3332354",

language = "English",

isbn = "978-1-4503-6269-6",

pages = "72--77",

booktitle = "IMIP '19: Proceedings of the 2019 International Conference on Intelligent Medicine and Image Processing",

publisher = "Association for Computing Machinery",

note = "3rd International Conference on Frontiers of Image Processing, ICFIP 2019 ; Conference date: 16-03-2019 Through 18-03-2019",

url = "http://www.icfip.org/",

}

Kolakalur, A & Vuksanovic, B 2019, Iterative reconstruction via preserved structures approach for CT images with limited scan angles. in IMIP '19: Proceedings of the 2019 International Conference on Intelligent Medicine and Image Processing. Association for Computing Machinery, pp. 72-77, 3rd International Conference on Frontiers of Image Processing, Florence, Italy, 16/03/19. https://doi.org/10.1145/3332340.3332354

Iterative reconstruction via preserved structures approach for CT images with limited scan angles. / Kolakalur, Anush; Vuksanovic, Branislav.
IMIP '19: Proceedings of the 2019 International Conference on Intelligent Medicine and Image Processing. Association for Computing Machinery, 2019. p. 72-77.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Iterative reconstruction via preserved structures approach for CT images with limited scan angles

AU - Kolakalur, Anush

AU - Vuksanovic, Branislav

PY - 2019/4/22

Y1 - 2019/4/22

N2 - Iterative image reconstruction techniques for computer tomography (CT) are finite iterations of forward-projections and backward projections. One of the major concerns related to this method is deterioration of the reconstructed images due to various image structure deformations during this procedure. This is usually manifested by blotchy and pixelated appearances of the reconstructed image with the effects becoming more pronounced for low and ultra-low scan angles. This paper proposes a new approach for the reconstruction of CT images ensuring the preservation of structural details and reduced image deterioration and deformation. We call this method iterative reconstruction through preserved structures (IR-PS). The results achieved using proposed IR-PS method are evaluated via RMSE (Root Mean Square Error) measure and SSIM (Structural SIMiliarity) index suggesting improvement in quality of reconstructed images.

AB - Iterative image reconstruction techniques for computer tomography (CT) are finite iterations of forward-projections and backward projections. One of the major concerns related to this method is deterioration of the reconstructed images due to various image structure deformations during this procedure. This is usually manifested by blotchy and pixelated appearances of the reconstructed image with the effects becoming more pronounced for low and ultra-low scan angles. This paper proposes a new approach for the reconstruction of CT images ensuring the preservation of structural details and reduced image deterioration and deformation. We call this method iterative reconstruction through preserved structures (IR-PS). The results achieved using proposed IR-PS method are evaluated via RMSE (Root Mean Square Error) measure and SSIM (Structural SIMiliarity) index suggesting improvement in quality of reconstructed images.

KW - computed tomography

KW - iterative image reconstruction

KW - image degradation

KW - limited scan angle

KW - CT reconstruction

UR - http://www.icfip.org/

U2 - 10.1145/3332340.3332354

DO - 10.1145/3332340.3332354

M3 - Conference contribution

SN - 978-1-4503-6269-6

SP - 72

EP - 77

BT - IMIP '19: Proceedings of the 2019 International Conference on Intelligent Medicine and Image Processing

PB - Association for Computing Machinery

T2 - 3rd International Conference on Frontiers of Image Processing

Y2 - 16 March 2019 through 18 March 2019

ER -

Iterative reconstruction via preserved structures approach for CT images with limited scan angles

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this