Convolutional neural network–based metal and streak artifacts reduction in dental CT images with sparse‐view sampling scheme

Seongjun Kim; Junhyun Ahn; Byeongjoon Kim; Chulhong Kim; Jongduk Baek

doi:10.1002/mp.15884

journal article Aug 10, 2022

Convolutional neural network–based metal and streak artifacts reduction in dental CT images with sparse‐view sampling scheme

Seongjun Kim Junhyun Ahn Byeongjoon Kim Chulhong Kim Jongduk Baek

Medical Physics Vol. 49 No. 9 pp. 6253-6277 · Wiley

View at Publisher Save 10.1002/mp.15884

Abstract

AbstractPurposeSparse‐view sampling has attracted attention for reducing the scan time and radiation dose of dental cone‐beam computed tomography (CBCT). Recently, various deep learning–based image reconstruction techniques for sparse‐view CT have been employed to produce high‐quality image while effectively reducing streak artifacts caused by the lack of projection views. However, most of these methods do not fully consider the effects of metal implants. As sparse‐view sampling strengthens the artifacts caused by metal objects, simultaneously reducing both metal and streak artifacts in sparse‐view CT images has been challenging. To solve this problem, in this study, we propose a novel framework.MethodsThe proposed method was based on the normalized metal artifact reduction (NMAR) method, and its performance was enhanced using two convolutional neural networks (CNNs). The first network reduced the initial artifacts while preserving the fine details to generate high‐quality priors for NMAR processing. Subsequently, the second network was employed to reduce the streak artifacts after NMAR processing of sparse‐view CT data. To validate the proposed method, we generated training and test data by computer simulations using both extended cardiac‐torso (XCAT) and clinical data sets.ResultsVisual inspection and quantitative evaluations demonstrated that the proposed method effectively reduced both metal and streak artifacts while preserving the details of anatomical structures compared with the conventional metal artifact reduction methods.ConclusionsWe propose a framework for reconstructing accurate CT images in metal‐inserted sparse‐view CT. The proposed method reduces streak artifacts from both metal objects and sparse‐view sampling while recovering the anatomical details, indicating the feasibility of fast‐scan dental CBCT imaging.

Topics

No keywords indexed for this article. Browse by subject →

References

64

[1]

10.1016/j.ejrad.2007.11.019

[2]

10.1007/s13244-014-0379-4

[3]

Borbély P "Orthodontic treatment of an impacted maxillary central incisor combined with surgical exposure" Management (2015)

[4]

Rigid and non-rigid motion artifact reduction in X-ray CT using attention module

Youngjun Ko, Seunghyuk Moon, Jongduk Baek et al.

Medical Image Analysis 10.1016/j.media.2020.101883

[5]

10.1016/j.oooo.2018.08.004

[6]

10.1002/mp.13731

[7]

10.1088/1361-6560/aa9254

[8]

10.1109/tmi.2018.2823338

[9]

Framing U-Net via Deep Convolutional Framelets: Application to Sparse-View CT

Yoseob Han, Jong Chul Ye

IEEE Transactions on Medical Imaging 10.1109/tmi.2018.2823768

[10]

Deep Convolutional Neural Network for Inverse Problems in Imaging

Kyong Hwan Jin, Michael T. McCann, Emmanuel Froustey et al.

IEEE Transactions on Image Processing 10.1109/tip.2017.2713099

[11]

LiangK YangH KangK XingY.Improve angular resolution for sparse‐view ct with residual convolutional neural network. InMedical Imaging 2018: Physics of Medical Imaging volume10573 page105731K.International Society for Optics and Photonics;2018. 10.1117/12.2293319

[12]

10.1002/mp.13258

[13]

10.1088/1361-6560/ab8fc1

[14]

LiangK YangH XingY.Comparison of projection domain image domain and comprehensive deep learning for sparse‐view x‐ray CT image reconstruction.arXiv preprint arXiv:1804.04289 2018.

[15]

10.1088/0031-9155/59/12/2997

[16]

Liu Y "Total variation‐stokes strategy for sparse‐view x‐ray CT image reconstruction" IEEE Trans Med Imaging (2013)

[17]

10.1088/0031-9155/57/23/7923

[18]

10.1109/cvpr42600.2020.00779

[19]

10.1088/0031-9155/60/19/7437

[20]

10.1109/tmi.2015.2478905

[21]

10.1118/1.1358303

[22]

10.1148/radiology.164.2.3602406

[23]

10.1118/1.3511507

[24]

10.1118/1.3484090

[25]

10.1109/tmi.2020.3005432

[26]

XuS DangH.Deep residual learning enabled metal artifact reduction in CT. InMedical Imaging 2018: Physics of Medical Imaging volume10573 page105733O.International Society for Optics and Photonics;2018. 10.1117/12.2293945

[27]

10.1117/12.2274427

[28]

Ghani MU "Deep learning based sinogram correction for metal artifact reduction" Electron Imaging (2018)

[29]

10.1002/mp.13199

[30]

10.1109/tmi.2018.2823083

[31]

10.1109/cvpr.2019.01076

[32]

10.1016/j.media.2021.102289

[33]

10.1117/1.jmi.8.5.052103

[34]

10.1186/1475-925x-11-44

[35]

10.23915/distill.00003

[36]

KingmaDP BaJ.Adam: a method for stochastic optimization.arXiv preprint arXiv:1412.6980 2014.

[37]

HubbellJH SeltzerSM.Tables of x‐ray mass attenuation coefficients and mass energy‐absorption coefficients from 1 kev to 20 mev for elements z = 1 to 92 and 48 additional substances of dosimetric interest.Tech Rep NIST.1995Jan 1 [revised 2017 Feb 17 10.6028/nist.ir.5632

[38]

cited 2020 Nov 9]. Accessed August 2020. available from:https://www.nist.gov/publications/tables‐x‐ray‐mass‐attenuation‐coefficientsand‐mass‐energy‐absorption‐coefficients‐1

[39]

The cancer imaging archive. Accessed April 8 2020.https://wiki.cancerimagingarchive.net/display/public/cptac‐hnscc

[40]

The cancer imaging archive. Accessed April 8 2020.https://wiki.cancerimagingarchive.net/display/public/tcga‐hnsc

[41]

10.1118/1.595715

[42]

Image quality assessment: from error visibility to structural similarity

Zhou Wang, A.C. Bovik, H.R. Sheikh et al.

IEEE Transactions on Image Processing 10.1109/tip.2003.819861

[43]

Low-Dose CT Image Denoising Using a Generative Adversarial Network With Wasserstein Distance and Perceptual Loss

Qingsong Yang, Pingkun Yan, Yanbo Zhang et al.

IEEE Transactions on Medical Imaging 10.1109/tmi.2018.2827462

[44]

10.1002/mp.13713

[45]

10.1016/j.compbiomed.2009.11.020

[46]

10.1155/2015/810796

[47]

10.1109/access.2020.2975826

[48]

10.1118/1.3691902

[49]

10.1007/978-3-030-00889-5_1

[50]

OktayO SchlemperJ FolgocLL et al.Attention U‐NET: learning where to look for the pancreas.arXiv preprint arXiv:1804.03999 2018.

Showing 50 of 64 references

Metrics

24

Citations

64

References

Details

Published: Aug 10, 2022
Vol/Issue: 49(9)
Pages: 6253-6277
License: View

Authors

S

Seongjun Kim

School of Integrated Technology Yonsei University Incheon South Korea

J

Junhyun Ahn

School of Integrated Technology Yonsei University Incheon South Korea

B

Byeongjoon Kim

School of Integrated Technology Yonsei University Incheon South Korea

C

Chulhong Kim

Departments of Electrical Engineering Convergence IT Engineering, Mechanical Engineering School of Interdisciplinary Bioscience and Bioengineering, and Medical Device Innovation Center Pohang University of Science and Technology Pohang South Korea

J

Jongduk Baek

School of Integrated Technology Yonsei University Incheon South Korea

Cite This Article

Seongjun Kim, Junhyun Ahn, Byeongjoon Kim, et al. (2022). Convolutional neural network–based metal and streak artifacts reduction in dental CT images with sparse‐view sampling scheme. Medical Physics, 49(9), 6253-6277. https://doi.org/10.1002/mp.15884

Convolutional neural network–based metal and streak artifacts reduction in dental CT images with sparse‐view sampling scheme

You May Also Like