A New Iterative Method for CT Reconstruction with Uncertain View Angles

Nicolai André Brogaard Riis; Yiqiu Dong

doi:10.1007/978-3-030-22368-7_13

A New Iterative Method for CT Reconstruction with Uncertain View Angles

Nicolai André Brogaard Riis^*, Yiqiu Dong

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

1 Downloads (Pure)

Abstract

In this paper, we propose a new iterative algorithm for Computed Tomography (CT) reconstruction when the problem has uncertainty in the view angles. The algorithm models this uncertainty by an additive model-discrepancy term leading to an estimate of the uncertainty in the likelihood function. This means we can combine state-of-the-art regularization priors such as total variation with this likelihood. To achieve a good reconstruction the algorithm alternates between updating the CT image and the uncertainty estimate in the likelihood. In simulated 2D numerical experiments, we show that our method is able to improve the relative reconstruction error and visual quality of the CT image for the uncertain-angle CT problem.

Original language	English
Title of host publication	Proceedings of 7th International Conference on Scale Space and Variational Methods in Computer Vision
Editors	Jan Lellmann, Jan Modersitzki, Martin Burger
Publisher	Springer
Publication date	1 Jan 2019
Pages	156-167
ISBN (Print)	9783030223670
DOIs	https://doi.org/10.1007/978-3-030-22368-7_13
Publication status	Published - 1 Jan 2019
Event	7th International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2019 - Hofgeismar, Germany Duration: 30 Jun 2019 → 4 Jul 2019

Conference

Conference	7th International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2019
Country	Germany
City	Hofgeismar
Period	30/06/2019 → 04/07/2019

Series	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11603 LNCS
ISSN	0302-9743

Keywords

Computed Tomography
Model discrepancy
Model error
Total variation
Uncertain view angles
Variational methods

Cite this

Riis, N. A. B., & Dong, Y. (2019). A New Iterative Method for CT Reconstruction with Uncertain View Angles. In J. Lellmann, J. Modersitzki, & M. Burger (Eds.), Proceedings of 7th International Conference on Scale Space and Variational Methods in Computer Vision (pp. 156-167). Springer. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol.. 11603 LNCS https://doi.org/10.1007/978-3-030-22368-7_13

Riis, Nicolai André Brogaard ; Dong, Yiqiu. / A New Iterative Method for CT Reconstruction with Uncertain View Angles. Proceedings of 7th International Conference on Scale Space and Variational Methods in Computer Vision. editor / Jan Lellmann ; Jan Modersitzki ; Martin Burger. Springer, 2019. pp. 156-167 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11603 LNCS).

@inproceedings{bf8e185640a447ae9ec7afcd418b2863,

title = "A New Iterative Method for CT Reconstruction with Uncertain View Angles",

abstract = "In this paper, we propose a new iterative algorithm for Computed Tomography (CT) reconstruction when the problem has uncertainty in the view angles. The algorithm models this uncertainty by an additive model-discrepancy term leading to an estimate of the uncertainty in the likelihood function. This means we can combine state-of-the-art regularization priors such as total variation with this likelihood. To achieve a good reconstruction the algorithm alternates between updating the CT image and the uncertainty estimate in the likelihood. In simulated 2D numerical experiments, we show that our method is able to improve the relative reconstruction error and visual quality of the CT image for the uncertain-angle CT problem.",

keywords = "Computed Tomography, Model discrepancy, Model error, Total variation, Uncertain view angles, Variational methods",

author = "Riis, {Nicolai Andr{\'e} Brogaard} and Yiqiu Dong",

year = "2019",

month = "1",

day = "1",

doi = "10.1007/978-3-030-22368-7_13",

language = "English",

isbn = "9783030223670",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "156--167",

editor = "Jan Lellmann and Jan Modersitzki and Martin Burger",

booktitle = "Proceedings of 7th International Conference on Scale Space and Variational Methods in Computer Vision",

}

Riis, NAB & Dong, Y 2019, A New Iterative Method for CT Reconstruction with Uncertain View Angles. in J Lellmann, J Modersitzki & M Burger (eds), Proceedings of 7th International Conference on Scale Space and Variational Methods in Computer Vision. Springer, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11603 LNCS, pp. 156-167, 7th International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2019, Hofgeismar, Germany, 30/06/2019. https://doi.org/10.1007/978-3-030-22368-7_13

A New Iterative Method for CT Reconstruction with Uncertain View Angles. / Riis, Nicolai André Brogaard ; Dong, Yiqiu.

Proceedings of 7th International Conference on Scale Space and Variational Methods in Computer Vision. ed. / Jan Lellmann; Jan Modersitzki; Martin Burger. Springer, 2019. p. 156-167 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11603 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

TY - GEN

T1 - A New Iterative Method for CT Reconstruction with Uncertain View Angles

AU - Riis, Nicolai André Brogaard

AU - Dong, Yiqiu

PY - 2019/1/1

Y1 - 2019/1/1

N2 - In this paper, we propose a new iterative algorithm for Computed Tomography (CT) reconstruction when the problem has uncertainty in the view angles. The algorithm models this uncertainty by an additive model-discrepancy term leading to an estimate of the uncertainty in the likelihood function. This means we can combine state-of-the-art regularization priors such as total variation with this likelihood. To achieve a good reconstruction the algorithm alternates between updating the CT image and the uncertainty estimate in the likelihood. In simulated 2D numerical experiments, we show that our method is able to improve the relative reconstruction error and visual quality of the CT image for the uncertain-angle CT problem.

AB - In this paper, we propose a new iterative algorithm for Computed Tomography (CT) reconstruction when the problem has uncertainty in the view angles. The algorithm models this uncertainty by an additive model-discrepancy term leading to an estimate of the uncertainty in the likelihood function. This means we can combine state-of-the-art regularization priors such as total variation with this likelihood. To achieve a good reconstruction the algorithm alternates between updating the CT image and the uncertainty estimate in the likelihood. In simulated 2D numerical experiments, we show that our method is able to improve the relative reconstruction error and visual quality of the CT image for the uncertain-angle CT problem.

KW - Computed Tomography

KW - Model discrepancy

KW - Model error

KW - Total variation

KW - Uncertain view angles

KW - Variational methods

U2 - 10.1007/978-3-030-22368-7_13

DO - 10.1007/978-3-030-22368-7_13

M3 - Article in proceedings

AN - SCOPUS:85068451193

SN - 9783030223670

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

SP - 156

EP - 167

BT - Proceedings of 7th International Conference on Scale Space and Variational Methods in Computer Vision

A2 - Lellmann, Jan

A2 - Modersitzki, Jan

A2 - Burger, Martin

PB - Springer

ER -

Riis NAB , Dong Y. A New Iterative Method for CT Reconstruction with Uncertain View Angles. In Lellmann J, Modersitzki J, Burger M, editors, Proceedings of 7th International Conference on Scale Space and Variational Methods in Computer Vision. Springer. 2019. p. 156-167. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Vol. 11603 LNCS). https://doi.org/10.1007/978-3-030-22368-7_13

Access to Document

10.1007/978-3-030-22368-7_13