Completion of Metal-Damaged Traces Based on Deep Learning in Sinogram Domain for Metal Artifacts Reduction in CT Images
In computed tomography (CT) images, the presence of metal artifacts leads to contaminated object structures. Theoretically, eliminating metal artifacts in the sinogram domain can correct projection deviation and provide reconstructed images that are more real. Contemporary methods that use deep netw...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-12-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/24/8164 |
_version_ | 1797500890088734720 |
---|---|
author | Linlin Zhu Yu Han Xiaoqi Xi Lei Li Bin Yan |
author_facet | Linlin Zhu Yu Han Xiaoqi Xi Lei Li Bin Yan |
author_sort | Linlin Zhu |
collection | DOAJ |
description | In computed tomography (CT) images, the presence of metal artifacts leads to contaminated object structures. Theoretically, eliminating metal artifacts in the sinogram domain can correct projection deviation and provide reconstructed images that are more real. Contemporary methods that use deep networks for completing metal-damaged sinogram data are limited to discontinuity at the boundaries of traces, which, however, lead to secondary artifacts. This study modifies the traditional U-net and adds two sinogram feature losses of projection images—namely, continuity and consistency of projection data at each angle, improving the accuracy of the complemented sinogram data. Masking the metal traces also ensures the stability and reliability of the unaffected data during metal artifacts reduction. The projection and reconstruction results and various evaluation metrics reveal that the proposed method can accurately repair missing data and reduce metal artifacts in reconstructed CT images. |
first_indexed | 2024-03-10T03:10:22Z |
format | Article |
id | doaj.art-20ffc97e2dcc4387b7b9d6514ee366f7 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-10T03:10:22Z |
publishDate | 2021-12-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-20ffc97e2dcc4387b7b9d6514ee366f72023-11-23T10:27:38ZengMDPI AGSensors1424-82202021-12-012124816410.3390/s21248164Completion of Metal-Damaged Traces Based on Deep Learning in Sinogram Domain for Metal Artifacts Reduction in CT ImagesLinlin Zhu0Yu Han1Xiaoqi Xi2Lei Li3Bin Yan4Henan Key Laboratory of Imaging and Intelligent Processing, PLA Strategic Support Force Information Engineering University, Zhengzhou 450001, ChinaHenan Key Laboratory of Imaging and Intelligent Processing, PLA Strategic Support Force Information Engineering University, Zhengzhou 450001, ChinaHenan Key Laboratory of Imaging and Intelligent Processing, PLA Strategic Support Force Information Engineering University, Zhengzhou 450001, ChinaHenan Key Laboratory of Imaging and Intelligent Processing, PLA Strategic Support Force Information Engineering University, Zhengzhou 450001, ChinaHenan Key Laboratory of Imaging and Intelligent Processing, PLA Strategic Support Force Information Engineering University, Zhengzhou 450001, ChinaIn computed tomography (CT) images, the presence of metal artifacts leads to contaminated object structures. Theoretically, eliminating metal artifacts in the sinogram domain can correct projection deviation and provide reconstructed images that are more real. Contemporary methods that use deep networks for completing metal-damaged sinogram data are limited to discontinuity at the boundaries of traces, which, however, lead to secondary artifacts. This study modifies the traditional U-net and adds two sinogram feature losses of projection images—namely, continuity and consistency of projection data at each angle, improving the accuracy of the complemented sinogram data. Masking the metal traces also ensures the stability and reliability of the unaffected data during metal artifacts reduction. The projection and reconstruction results and various evaluation metrics reveal that the proposed method can accurately repair missing data and reduce metal artifacts in reconstructed CT images.https://www.mdpi.com/1424-8220/21/24/8164computed tomographymetal artifacts reductiondeep learningsinogram |
spellingShingle | Linlin Zhu Yu Han Xiaoqi Xi Lei Li Bin Yan Completion of Metal-Damaged Traces Based on Deep Learning in Sinogram Domain for Metal Artifacts Reduction in CT Images Sensors computed tomography metal artifacts reduction deep learning sinogram |
title | Completion of Metal-Damaged Traces Based on Deep Learning in Sinogram Domain for Metal Artifacts Reduction in CT Images |
title_full | Completion of Metal-Damaged Traces Based on Deep Learning in Sinogram Domain for Metal Artifacts Reduction in CT Images |
title_fullStr | Completion of Metal-Damaged Traces Based on Deep Learning in Sinogram Domain for Metal Artifacts Reduction in CT Images |
title_full_unstemmed | Completion of Metal-Damaged Traces Based on Deep Learning in Sinogram Domain for Metal Artifacts Reduction in CT Images |
title_short | Completion of Metal-Damaged Traces Based on Deep Learning in Sinogram Domain for Metal Artifacts Reduction in CT Images |
title_sort | completion of metal damaged traces based on deep learning in sinogram domain for metal artifacts reduction in ct images |
topic | computed tomography metal artifacts reduction deep learning sinogram |
url | https://www.mdpi.com/1424-8220/21/24/8164 |
work_keys_str_mv | AT linlinzhu completionofmetaldamagedtracesbasedondeeplearninginsinogramdomainformetalartifactsreductioninctimages AT yuhan completionofmetaldamagedtracesbasedondeeplearninginsinogramdomainformetalartifactsreductioninctimages AT xiaoqixi completionofmetaldamagedtracesbasedondeeplearninginsinogramdomainformetalartifactsreductioninctimages AT leili completionofmetaldamagedtracesbasedondeeplearninginsinogramdomainformetalartifactsreductioninctimages AT binyan completionofmetaldamagedtracesbasedondeeplearninginsinogramdomainformetalartifactsreductioninctimages |