View in EDS

Task-specific self-supervision for CT image denoising.

Saved in:
Bibliographic Details
Title: Task-specific self-supervision for CT image denoising.
Authors: Haque, Ayaan1,2 (AUTHOR), Wang, Adam2 (AUTHOR), Al Zubaer Imran, Abdullah2,3 (AUTHOR) aimran@uky.edu
Source: Computer Methods in Biomechanics & Biomedical Engineering: Imaging & Visualisation. Dec2023, Vol. 11 Issue 6, p2196-2208. 13p.
Subjects: Image denoising, Computed tomography, Blended learning, Deep learning, Radiation doses
Abstract: CT image quality is largely reliant on radiation dose, which causes a trade-off between image quality and dose, affecting the subsequent image-based diagnostic and treatment performances. Deep learning approaches for low-dose CT image denoising require access to large training sets and specifically the reference full-dose images, which can be difficult to obtain. Self-supervised learning enables learning with reduced reference data burden. Currently available self-supervised CT denoising works are either dependent on foreign domains or pretexts that are not very task-relevant, requiring lots of adjustments (architectural, loss function, training parameters, etc.) and additional skills to perform the downstream denoising task. To tackle the aforementioned challenges, we propose a novel self-supervised pretraining approach, namely Self-Supervised Window-Leveling for Image DeNoising (SSWL-IDN), leveraging an innovative, task-relevant, simple yet effective surrogate – prediction of the window-leveled equivalent. SSWL-IDN leverages residual learning and a hybrid loss combining perceptual loss and MSE, all incorporated in a VAE framework. Our extensive (in- and cross-domain) experimentation demonstrates the effectiveness of SSWL-IDN in aggressive denoising of CT (abdomen and chest) images acquired at two different dose levels (5% and 25%), without requiring any kind of architectural or downstream training adjustments. We have made our code publicly available at [ABSTRACT FROM AUTHOR]
Copyright of Computer Methods in Biomechanics & Biomedical Engineering: Imaging & Visualisation is the property of Taylor & Francis Ltd 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
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 174632733
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Task-specific self-supervision for CT image denoising.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Haque%2C+Ayaan%22">Haque, Ayaan</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Wang%2C+Adam%22">Wang, Adam</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Al+Zubaer+Imran%2C+Abdullah%22">Al Zubaer Imran, Abdullah</searchLink><relatesTo>2,3</relatesTo> (AUTHOR)<i> aimran@uky.edu</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Computer+Methods+in+Biomechanics+%26+Biomedical+Engineering%3A+Imaging+%26+Visualisation%22">Computer Methods in Biomechanics & Biomedical Engineering: Imaging & Visualisation</searchLink>. Dec2023, Vol. 11 Issue 6, p2196-2208. 13p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Image+denoising%22">Image denoising</searchLink><br /><searchLink fieldCode="DE" term="%22Computed+tomography%22">Computed tomography</searchLink><br /><searchLink fieldCode="DE" term="%22Blended+learning%22">Blended learning</searchLink><br /><searchLink fieldCode="DE" term="%22Deep+learning%22">Deep learning</searchLink><br /><searchLink fieldCode="DE" term="%22Radiation+doses%22">Radiation doses</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: CT image quality is largely reliant on radiation dose, which causes a trade-off between image quality and dose, affecting the subsequent image-based diagnostic and treatment performances. Deep learning approaches for low-dose CT image denoising require access to large training sets and specifically the reference full-dose images, which can be difficult to obtain. Self-supervised learning enables learning with reduced reference data burden. Currently available self-supervised CT denoising works are either dependent on foreign domains or pretexts that are not very task-relevant, requiring lots of adjustments (architectural, loss function, training parameters, etc.) and additional skills to perform the downstream denoising task. To tackle the aforementioned challenges, we propose a novel self-supervised pretraining approach, namely Self-Supervised Window-Leveling for Image DeNoising (SSWL-IDN), leveraging an innovative, task-relevant, simple yet effective surrogate – prediction of the window-leveled equivalent. SSWL-IDN leverages residual learning and a hybrid loss combining perceptual loss and MSE, all incorporated in a VAE framework. Our extensive (in- and cross-domain) experimentation demonstrates the effectiveness of SSWL-IDN in aggressive denoising of CT (abdomen and chest) images acquired at two different dose levels (5% and 25%), without requiring any kind of architectural or downstream training adjustments. We have made our code publicly available at [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Computer Methods in Biomechanics & Biomedical Engineering: Imaging & Visualisation is the property of Taylor & Francis Ltd 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=174632733
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/21681163.2023.2219771
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 13
        StartPage: 2196
    Subjects:
      – SubjectFull: Image denoising
        Type: general
      – SubjectFull: Computed tomography
        Type: general
      – SubjectFull: Blended learning
        Type: general
      – SubjectFull: Deep learning
        Type: general
      – SubjectFull: Radiation doses
        Type: general
    Titles:
      – TitleFull: Task-specific self-supervision for CT image denoising.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Haque, Ayaan
      – PersonEntity:
          Name:
            NameFull: Wang, Adam
      – PersonEntity:
          Name:
            NameFull: Al Zubaer Imran, Abdullah
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 12
              Text: Dec2023
              Type: published
              Y: 2023
          Identifiers:
            – Type: issn-print
              Value: 21681163
          Numbering:
            – Type: volume
              Value: 11
            – Type: issue
              Value: 6
          Titles:
            – TitleFull: Computer Methods in Biomechanics & Biomedical Engineering: Imaging & Visualisation
              Type: main
ResultId 1