View in EDS HTML Full Text PDF Full Text

Mitigation of motion‐induced artifacts in cone beam computed tomography using deep convolutional neural networks.

Saved in:
Bibliographic Details
Title: Mitigation of motion‐induced artifacts in cone beam computed tomography using deep convolutional neural networks.
Authors: Amirian, Mohammadreza1,2 (AUTHOR), Montoya‐Zegarra, Javier A.1 (AUTHOR), Herzig, Ivo3 (AUTHOR), Eggenberger Hotz, Peter3 (AUTHOR), Lichtensteiger, Lukas3 (AUTHOR), Morf, Marco3 (AUTHOR), Züst, Alexander3 (AUTHOR), Paysan, Pascal4 (AUTHOR), Peterlik, Igor4 (AUTHOR), Scheib, Stefan4 (AUTHOR), Füchslin, Rudolf Marcel3,5 (AUTHOR), Stadelmann, Thilo1,5 (AUTHOR), Schilling, Frank‐Peter1 (AUTHOR) scik@zhaw.ch
Source: Medical Physics. Oct2023, Vol. 50 Issue 10, p6228-6242. 15p.
Subjects: Artificial neural networks, Convolutional neural networks, Cone beam computed tomography, Image reconstruction algorithms, Image-guided radiation therapy, Nomography (Mathematics), Supervised learning, Computed tomography
Abstract: Background: Cone beam computed tomography (CBCT) is often employed on radiation therapy treatment devices (linear accelerators) used in image‐guided radiation therapy (IGRT). For each treatment session, it is necessary to obtain the image of the day in order to accurately position the patient and to enable adaptive treatment capabilities including auto‐segmentation and dose calculation. Reconstructed CBCT images often suffer from artifacts, in particular those induced by patient motion. Deep‐learning based approaches promise ways to mitigate such artifacts. Purpose: We propose a novel deep‐learning based approach with the goal to reduce motion induced artifacts in CBCT images and improve image quality. It is based on supervised learning and includes neural network architectures employed as pre‐ and/or post‐processing steps during CBCT reconstruction. Methods: Our approach is based on deep convolutional neural networks which complement the standard CBCT reconstruction, which is performed either with the analytical Feldkamp‐Davis‐Kress (FDK) method, or with an iterative algebraic reconstruction technique (SART‐TV). The neural networks, which are based on refined U‐net architectures, are trained end‐to‐end in a supervised learning setup. Labeled training data are obtained by means of a motion simulation, which uses the two extreme phases of 4D CT scans, their deformation vector fields, as well as time‐dependent amplitude signals as input. The trained networks are validated against ground truth using quantitative metrics, as well as by using real patient CBCT scans for a qualitative evaluation by clinical experts. Results: The presented novel approach is able to generalize to unseen data and yields significant reductions in motion induced artifacts as well as improvements in image quality compared with existing state‐of‐the‐art CBCT reconstruction algorithms (up to +6.3 dB and +0.19 improvements in peak signal‐to‐noise ratio, PSNR, and structural similarity index measure, SSIM, respectively), as evidenced by validation with an unseen test dataset, and confirmed by a clinical evaluation on real patient scans (up to 74% preference for motion artifact reduction over standard reconstruction). Conclusions: For the first time, it is demonstrated, also by means of clinical evaluation, that inserting deep neural networks as pre‐ and post‐processing plugins in the existing 3D CBCT reconstruction and trained end‐to‐end yield significant improvements in image quality and reduction of motion artifacts. [ABSTRACT FROM AUTHOR]
Copyright of Medical Physics is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: egs
DbLabel: Engineering Source
An: 172913945
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Mitigation of motion‐induced artifacts in cone beam computed tomography using deep convolutional neural networks.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Amirian%2C+Mohammadreza%22">Amirian, Mohammadreza</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Montoya‐Zegarra%2C+Javier+A%2E%22">Montoya‐Zegarra, Javier A.</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Herzig%2C+Ivo%22">Herzig, Ivo</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Eggenberger+Hotz%2C+Peter%22">Eggenberger Hotz, Peter</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Lichtensteiger%2C+Lukas%22">Lichtensteiger, Lukas</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Morf%2C+Marco%22">Morf, Marco</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Züst%2C+Alexander%22">Züst, Alexander</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Paysan%2C+Pascal%22">Paysan, Pascal</searchLink><relatesTo>4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Peterlik%2C+Igor%22">Peterlik, Igor</searchLink><relatesTo>4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Scheib%2C+Stefan%22">Scheib, Stefan</searchLink><relatesTo>4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Füchslin%2C+Rudolf+Marcel%22">Füchslin, Rudolf Marcel</searchLink><relatesTo>3,5</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Stadelmann%2C+Thilo%22">Stadelmann, Thilo</searchLink><relatesTo>1,5</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Schilling%2C+Frank‐Peter%22">Schilling, Frank‐Peter</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> scik@zhaw.ch</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Medical+Physics%22">Medical Physics</searchLink>. Oct2023, Vol. 50 Issue 10, p6228-6242. 15p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Artificial+neural+networks%22">Artificial neural networks</searchLink><br /><searchLink fieldCode="DE" term="%22Convolutional+neural+networks%22">Convolutional neural networks</searchLink><br /><searchLink fieldCode="DE" term="%22Cone+beam+computed+tomography%22">Cone beam computed tomography</searchLink><br /><searchLink fieldCode="DE" term="%22Image+reconstruction+algorithms%22">Image reconstruction algorithms</searchLink><br /><searchLink fieldCode="DE" term="%22Image-guided+radiation+therapy%22">Image-guided radiation therapy</searchLink><br /><searchLink fieldCode="DE" term="%22Nomography+%28Mathematics%29%22">Nomography (Mathematics)</searchLink><br /><searchLink fieldCode="DE" term="%22Supervised+learning%22">Supervised learning</searchLink><br /><searchLink fieldCode="DE" term="%22Computed+tomography%22">Computed tomography</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Background: Cone beam computed tomography (CBCT) is often employed on radiation therapy treatment devices (linear accelerators) used in image‐guided radiation therapy (IGRT). For each treatment session, it is necessary to obtain the image of the day in order to accurately position the patient and to enable adaptive treatment capabilities including auto‐segmentation and dose calculation. Reconstructed CBCT images often suffer from artifacts, in particular those induced by patient motion. Deep‐learning based approaches promise ways to mitigate such artifacts. Purpose: We propose a novel deep‐learning based approach with the goal to reduce motion induced artifacts in CBCT images and improve image quality. It is based on supervised learning and includes neural network architectures employed as pre‐ and/or post‐processing steps during CBCT reconstruction. Methods: Our approach is based on deep convolutional neural networks which complement the standard CBCT reconstruction, which is performed either with the analytical Feldkamp‐Davis‐Kress (FDK) method, or with an iterative algebraic reconstruction technique (SART‐TV). The neural networks, which are based on refined U‐net architectures, are trained end‐to‐end in a supervised learning setup. Labeled training data are obtained by means of a motion simulation, which uses the two extreme phases of 4D CT scans, their deformation vector fields, as well as time‐dependent amplitude signals as input. The trained networks are validated against ground truth using quantitative metrics, as well as by using real patient CBCT scans for a qualitative evaluation by clinical experts. Results: The presented novel approach is able to generalize to unseen data and yields significant reductions in motion induced artifacts as well as improvements in image quality compared with existing state‐of‐the‐art CBCT reconstruction algorithms (up to +6.3 dB and +0.19 improvements in peak signal‐to‐noise ratio, PSNR, and structural similarity index measure, SSIM, respectively), as evidenced by validation with an unseen test dataset, and confirmed by a clinical evaluation on real patient scans (up to 74% preference for motion artifact reduction over standard reconstruction). Conclusions: For the first time, it is demonstrated, also by means of clinical evaluation, that inserting deep neural networks as pre‐ and post‐processing plugins in the existing 3D CBCT reconstruction and trained end‐to‐end yield significant improvements in image quality and reduction of motion artifacts. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Medical Physics is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=172913945
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1002/mp.16405
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 15
        StartPage: 6228
    Subjects:
      – SubjectFull: Artificial neural networks
        Type: general
      – SubjectFull: Convolutional neural networks
        Type: general
      – SubjectFull: Cone beam computed tomography
        Type: general
      – SubjectFull: Image reconstruction algorithms
        Type: general
      – SubjectFull: Image-guided radiation therapy
        Type: general
      – SubjectFull: Nomography (Mathematics)
        Type: general
      – SubjectFull: Supervised learning
        Type: general
      – SubjectFull: Computed tomography
        Type: general
    Titles:
      – TitleFull: Mitigation of motion‐induced artifacts in cone beam computed tomography using deep convolutional neural networks.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Amirian, Mohammadreza
      – PersonEntity:
          Name:
            NameFull: Montoya‐Zegarra, Javier A.
      – PersonEntity:
          Name:
            NameFull: Herzig, Ivo
      – PersonEntity:
          Name:
            NameFull: Eggenberger Hotz, Peter
      – PersonEntity:
          Name:
            NameFull: Lichtensteiger, Lukas
      – PersonEntity:
          Name:
            NameFull: Morf, Marco
      – PersonEntity:
          Name:
            NameFull: Züst, Alexander
      – PersonEntity:
          Name:
            NameFull: Paysan, Pascal
      – PersonEntity:
          Name:
            NameFull: Peterlik, Igor
      – PersonEntity:
          Name:
            NameFull: Scheib, Stefan
      – PersonEntity:
          Name:
            NameFull: Füchslin, Rudolf Marcel
      – PersonEntity:
          Name:
            NameFull: Stadelmann, Thilo
      – PersonEntity:
          Name:
            NameFull: Schilling, Frank‐Peter
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 10
              Text: Oct2023
              Type: published
              Y: 2023
          Identifiers:
            – Type: issn-print
              Value: 00942405
          Numbering:
            – Type: volume
              Value: 50
            – Type: issue
              Value: 10
          Titles:
            – TitleFull: Medical Physics
              Type: main
ResultId 1