Correction of the Beam Hardening Artifacts in CT Images Using Pix2pixGAN Network

Legeng Lin; Shunli Wang; Zhisheng Wang; Zihan Deng; Zongfeng Li; Junning Cui

doi:10.1109/ICEMI59194.2023.10270262

Correction of the Beam Hardening Artifacts in CT Images Using Pix2pixGAN Network

Legeng Lin
, Shunli Wang
, Zhisheng Wang
, Zihan Deng
, Zongfeng Li
, Junning Cui^*

^*Corresponding author for this work

School of Instrumentation Science and Engineering

Center of Ultra-precision Optoelectronic Instrument Engeering
Harbin Institute of Technology
Ministry of Industry and Infornation Technolgy

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

1 Scopus citations

Abstract

Due to the noncoincidence between the mono energetic rays-based reconstruction algorithms and the multi-energy projection data, there always exit beam hardening artifacts destroying the reconstructed CT images. A correction of the beam hardening artifacts can effectively enhance the accuracy and reliability of the scan results. In this preliminary study, we propose a pix2pixGAN generative adversarial network to do this work. Compared to the traditional approaches, this method can increase the structural similarity index measure by 11.81%, improve the decrease the peak signal-to-noise ratio by 25.59% and decrease the root-mean-square error by 62.02%. These results demonstrate that the pix2pixGAN generative adversarial network can simultaneously correct the artifacts and improve the recovery of details.

Original language	English
Title of host publication	Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023
Editors	Juan Wu, Jiali Yin
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	60-64
Number of pages	5
ISBN (Electronic)	9798350327144
DOIs	https://doi.org/10.1109/ICEMI59194.2023.10270262
State	Published - 2023
Event	16th IEEE International Conference on Electronic Measurement and Instruments, ICEMI 2023 - Harbin, China Duration: 9 Aug 2023 → 11 Aug 2023

Publication series

Name	Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023

Conference

Conference	16th IEEE International Conference on Electronic Measurement and Instruments, ICEMI 2023
Country/Territory	China
City	Harbin
Period	9/08/23 → 11/08/23

Keywords

beam hardening artifacts
computed tomography
deep learning network
pix2pixGAN generative adversarial network

Access to Document

10.1109/ICEMI59194.2023.10270262

Cite this

Lin, L., Wang, S., Wang, Z., Deng, Z., Li, Z., & Cui, J. (2023). Correction of the Beam Hardening Artifacts in CT Images Using Pix2pixGAN Network. In J. Wu, & J. Yin (Eds.), Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023 (pp. 60-64). (Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICEMI59194.2023.10270262

Lin, Legeng ; Wang, Shunli ; Wang, Zhisheng et al. / Correction of the Beam Hardening Artifacts in CT Images Using Pix2pixGAN Network. Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023. editor / Juan Wu ; Jiali Yin. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 60-64 (Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023).

@inproceedings{ac6a683c29a94dc5af55348638b78443,

title = "Correction of the Beam Hardening Artifacts in CT Images Using Pix2pixGAN Network",

abstract = "Due to the noncoincidence between the mono energetic rays-based reconstruction algorithms and the multi-energy projection data, there always exit beam hardening artifacts destroying the reconstructed CT images. A correction of the beam hardening artifacts can effectively enhance the accuracy and reliability of the scan results. In this preliminary study, we propose a pix2pixGAN generative adversarial network to do this work. Compared to the traditional approaches, this method can increase the structural similarity index measure by 11.81\%, improve the decrease the peak signal-to-noise ratio by 25.59\% and decrease the root-mean-square error by 62.02\%. These results demonstrate that the pix2pixGAN generative adversarial network can simultaneously correct the artifacts and improve the recovery of details.",

keywords = "beam hardening artifacts, computed tomography, deep learning network, pix2pixGAN generative adversarial network",

author = "Legeng Lin and Shunli Wang and Zhisheng Wang and Zihan Deng and Zongfeng Li and Junning Cui",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 16th IEEE International Conference on Electronic Measurement and Instruments, ICEMI 2023 ; Conference date: 09-08-2023 Through 11-08-2023",

year = "2023",

doi = "10.1109/ICEMI59194.2023.10270262",

language = "英语",

series = "Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "60--64",

editor = "Juan Wu and Jiali Yin",

booktitle = "Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023",

address = "美国",

}

Lin, L, Wang, S, Wang, Z, Deng, Z, Li, Z & Cui, J 2023, Correction of the Beam Hardening Artifacts in CT Images Using Pix2pixGAN Network. in J Wu & J Yin (eds), Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023. Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023, Institute of Electrical and Electronics Engineers Inc., pp. 60-64, 16th IEEE International Conference on Electronic Measurement and Instruments, ICEMI 2023, Harbin, China, 9/08/23. https://doi.org/10.1109/ICEMI59194.2023.10270262

Correction of the Beam Hardening Artifacts in CT Images Using Pix2pixGAN Network. / Lin, Legeng; Wang, Shunli; Wang, Zhisheng et al.
Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023. ed. / Juan Wu; Jiali Yin. Institute of Electrical and Electronics Engineers Inc., 2023. p. 60-64 (Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023).

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

TY - GEN

T1 - Correction of the Beam Hardening Artifacts in CT Images Using Pix2pixGAN Network

AU - Lin, Legeng

AU - Wang, Shunli

AU - Wang, Zhisheng

AU - Deng, Zihan

AU - Li, Zongfeng

AU - Cui, Junning

PY - 2023

Y1 - 2023

N2 - Due to the noncoincidence between the mono energetic rays-based reconstruction algorithms and the multi-energy projection data, there always exit beam hardening artifacts destroying the reconstructed CT images. A correction of the beam hardening artifacts can effectively enhance the accuracy and reliability of the scan results. In this preliminary study, we propose a pix2pixGAN generative adversarial network to do this work. Compared to the traditional approaches, this method can increase the structural similarity index measure by 11.81%, improve the decrease the peak signal-to-noise ratio by 25.59% and decrease the root-mean-square error by 62.02%. These results demonstrate that the pix2pixGAN generative adversarial network can simultaneously correct the artifacts and improve the recovery of details.

AB - Due to the noncoincidence between the mono energetic rays-based reconstruction algorithms and the multi-energy projection data, there always exit beam hardening artifacts destroying the reconstructed CT images. A correction of the beam hardening artifacts can effectively enhance the accuracy and reliability of the scan results. In this preliminary study, we propose a pix2pixGAN generative adversarial network to do this work. Compared to the traditional approaches, this method can increase the structural similarity index measure by 11.81%, improve the decrease the peak signal-to-noise ratio by 25.59% and decrease the root-mean-square error by 62.02%. These results demonstrate that the pix2pixGAN generative adversarial network can simultaneously correct the artifacts and improve the recovery of details.

KW - beam hardening artifacts

KW - computed tomography

KW - deep learning network

KW - pix2pixGAN generative adversarial network

UR - https://www.scopus.com/pages/publications/85174954207

U2 - 10.1109/ICEMI59194.2023.10270262

DO - 10.1109/ICEMI59194.2023.10270262

M3 - 会议稿件

AN - SCOPUS:85174954207

T3 - Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023

SP - 60

EP - 64

BT - Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023

A2 - Wu, Juan

A2 - Yin, Jiali

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 16th IEEE International Conference on Electronic Measurement and Instruments, ICEMI 2023

Y2 - 9 August 2023 through 11 August 2023

ER -

Lin L, Wang S, Wang Z, Deng Z, Li Z, Cui J. Correction of the Beam Hardening Artifacts in CT Images Using Pix2pixGAN Network. In Wu J, Yin J, editors, Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023. Institute of Electrical and Electronics Engineers Inc. 2023. p. 60-64. (Proceedings of 2023 IEEE 16th International Conference on Electronic Measurement and Instruments, ICEMI 2023). doi: 10.1109/ICEMI59194.2023.10270262

Correction of the Beam Hardening Artifacts in CT Images Using Pix2pixGAN Network

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this