View in EDS HTML Full Text PDF Full Text

Neural network algorithm for detection of erosions and ankylosis on CT of the sacroiliac joints: multicentre development and validation of diagnostic accuracy.

Saved in:
Bibliographic Details
Title: Neural network algorithm for detection of erosions and ankylosis on CT of the sacroiliac joints: multicentre development and validation of diagnostic accuracy.
Authors: Van Den Berghe, Thomas1 (AUTHOR) thovdnbe.vandenberghe@ugent.be, Babin, Danilo2 (AUTHOR), Chen, Min3 (AUTHOR), Callens, Martijn1 (AUTHOR), Brack, Denim1 (AUTHOR), Maes, Helena1 (AUTHOR), Lievens, Jan1 (AUTHOR), Lammens, Marie1 (AUTHOR), Van Sumere, Maxime1 (AUTHOR), Morbée, Lieve1 (AUTHOR), Hautekeete, Simon1 (AUTHOR), Schatteman, Stijn1 (AUTHOR), Jacobs, Tom1 (AUTHOR), Thooft, Willem-Jan1 (AUTHOR), Herregods, Nele1 (AUTHOR), Huysse, Wouter1 (AUTHOR), Jaremko, Jacob L.4 (AUTHOR), Lambert, Robert4 (AUTHOR), Maksymowych, Walter4 (AUTHOR), Laloo, Frederiek1 (AUTHOR)
Source: European Radiology. Nov2023, Vol. 33 Issue 11, p8310-8323. 14p.
Subjects: Ghent University, Sacroiliac joint, Ankylosis, Convolutional neural networks, Computed tomography, Deep learning
Abstract: Objectives: To evaluate the feasibility and diagnostic accuracy of a deep learning network for detection of structural lesions of sacroiliitis on multicentre pelvic CT scans. Methods: Pelvic CT scans of 145 patients (81 female, 121 Ghent University/24 Alberta University, 18–87 years old, mean 40 ± 13 years, 2005–2021) with a clinical suspicion of sacroiliitis were retrospectively included. After manual sacroiliac joint (SIJ) segmentation and structural lesion annotation, a U-Net for SIJ segmentation and two separate convolutional neural networks (CNN) for erosion and ankylosis detection were trained. In-training validation and tenfold validation testing (U-Net—n = 10 × 58; CNN—n = 10 × 29) on a test dataset were performed to assess performance on a slice-by-slice and patient level (dice coefficient/accuracy/sensitivity/specificity/positive and negative predictive value/ROC AUC). Patient-level optimisation was applied to increase the performance regarding predefined statistical metrics. Gradient-weighted class activation mapping (Grad-CAM++) heatmap explainability analysis highlighted image parts with statistically important regions for algorithmic decisions. Results: Regarding SIJ segmentation, a dice coefficient of 0.75 was obtained in the test dataset. For slice-by-slice structural lesion detection, a sensitivity/specificity/ROC AUC of 95%/89%/0.92 and 93%/91%/0.91 were obtained in the test dataset for erosion and ankylosis detection, respectively. For patient-level lesion detection after pipeline optimisation for predefined statistical metrics, a sensitivity/specificity of 95%/85% and 82%/97% were obtained for erosion and ankylosis detection, respectively. Grad-CAM++ explainability analysis highlighted cortical edges as focus for pipeline decisions. Conclusions: An optimised deep learning pipeline, including an explainability analysis, detects structural lesions of sacroiliitis on pelvic CT scans with excellent statistical performance on a slice-by-slice and patient level. Clinical relevance statement: An optimised deep learning pipeline, including a robust explainability analysis, detects structural lesions of sacroiliitis on pelvic CT scans with excellent statistical metrics on a slice-by-slice and patient level. Key Points: • Structural lesions of sacroiliitis can be detected automatically in pelvic CT scans. • Both automatic segmentation and disease detection yield excellent statistical outcome metrics. • The algorithm takes decisions based on cortical edges, rendering an explainable solution. [ABSTRACT FROM AUTHOR]
Copyright of European Radiology is the property of Springer Nature 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: 173151700
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Neural network algorithm for detection of erosions and ankylosis on CT of the sacroiliac joints: multicentre development and validation of diagnostic accuracy.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Van+Den+Berghe%2C+Thomas%22">Van Den Berghe, Thomas</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> thovdnbe.vandenberghe@ugent.be</i><br /><searchLink fieldCode="AR" term="%22Babin%2C+Danilo%22">Babin, Danilo</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Chen%2C+Min%22">Chen, Min</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Callens%2C+Martijn%22">Callens, Martijn</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Brack%2C+Denim%22">Brack, Denim</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Maes%2C+Helena%22">Maes, Helena</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Lievens%2C+Jan%22">Lievens, Jan</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Lammens%2C+Marie%22">Lammens, Marie</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Van+Sumere%2C+Maxime%22">Van Sumere, Maxime</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Morbée%2C+Lieve%22">Morbée, Lieve</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Hautekeete%2C+Simon%22">Hautekeete, Simon</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Schatteman%2C+Stijn%22">Schatteman, Stijn</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Jacobs%2C+Tom%22">Jacobs, Tom</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Thooft%2C+Willem-Jan%22">Thooft, Willem-Jan</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Herregods%2C+Nele%22">Herregods, Nele</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Huysse%2C+Wouter%22">Huysse, Wouter</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Jaremko%2C+Jacob+L%2E%22">Jaremko, Jacob L.</searchLink><relatesTo>4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Lambert%2C+Robert%22">Lambert, Robert</searchLink><relatesTo>4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Maksymowych%2C+Walter%22">Maksymowych, Walter</searchLink><relatesTo>4</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Laloo%2C+Frederiek%22">Laloo, Frederiek</searchLink><relatesTo>1</relatesTo> (AUTHOR)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22European+Radiology%22">European Radiology</searchLink>. Nov2023, Vol. 33 Issue 11, p8310-8323. 14p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Ghent+University%22">Ghent University</searchLink><br /><searchLink fieldCode="DE" term="%22Sacroiliac+joint%22">Sacroiliac joint</searchLink><br /><searchLink fieldCode="DE" term="%22Ankylosis%22">Ankylosis</searchLink><br /><searchLink fieldCode="DE" term="%22Convolutional+neural+networks%22">Convolutional neural networks</searchLink><br /><searchLink fieldCode="DE" term="%22Computed+tomography%22">Computed tomography</searchLink><br /><searchLink fieldCode="DE" term="%22Deep+learning%22">Deep learning</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Objectives: To evaluate the feasibility and diagnostic accuracy of a deep learning network for detection of structural lesions of sacroiliitis on multicentre pelvic CT scans. Methods: Pelvic CT scans of 145 patients (81 female, 121 Ghent University/24 Alberta University, 18–87 years old, mean 40 ± 13 years, 2005–2021) with a clinical suspicion of sacroiliitis were retrospectively included. After manual sacroiliac joint (SIJ) segmentation and structural lesion annotation, a U-Net for SIJ segmentation and two separate convolutional neural networks (CNN) for erosion and ankylosis detection were trained. In-training validation and tenfold validation testing (U-Net—n = 10 × 58; CNN—n = 10 × 29) on a test dataset were performed to assess performance on a slice-by-slice and patient level (dice coefficient/accuracy/sensitivity/specificity/positive and negative predictive value/ROC AUC). Patient-level optimisation was applied to increase the performance regarding predefined statistical metrics. Gradient-weighted class activation mapping (Grad-CAM++) heatmap explainability analysis highlighted image parts with statistically important regions for algorithmic decisions. Results: Regarding SIJ segmentation, a dice coefficient of 0.75 was obtained in the test dataset. For slice-by-slice structural lesion detection, a sensitivity/specificity/ROC AUC of 95%/89%/0.92 and 93%/91%/0.91 were obtained in the test dataset for erosion and ankylosis detection, respectively. For patient-level lesion detection after pipeline optimisation for predefined statistical metrics, a sensitivity/specificity of 95%/85% and 82%/97% were obtained for erosion and ankylosis detection, respectively. Grad-CAM++ explainability analysis highlighted cortical edges as focus for pipeline decisions. Conclusions: An optimised deep learning pipeline, including an explainability analysis, detects structural lesions of sacroiliitis on pelvic CT scans with excellent statistical performance on a slice-by-slice and patient level. Clinical relevance statement: An optimised deep learning pipeline, including a robust explainability analysis, detects structural lesions of sacroiliitis on pelvic CT scans with excellent statistical metrics on a slice-by-slice and patient level. Key Points: • Structural lesions of sacroiliitis can be detected automatically in pelvic CT scans. • Both automatic segmentation and disease detection yield excellent statistical outcome metrics. • The algorithm takes decisions based on cortical edges, rendering an explainable solution. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of European Radiology is the property of Springer Nature 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=173151700
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s00330-023-09704-y
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 14
        StartPage: 8310
    Subjects:
      – SubjectFull: Ghent University
        Type: general
      – SubjectFull: Sacroiliac joint
        Type: general
      – SubjectFull: Ankylosis
        Type: general
      – SubjectFull: Convolutional neural networks
        Type: general
      – SubjectFull: Computed tomography
        Type: general
      – SubjectFull: Deep learning
        Type: general
    Titles:
      – TitleFull: Neural network algorithm for detection of erosions and ankylosis on CT of the sacroiliac joints: multicentre development and validation of diagnostic accuracy.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Van Den Berghe, Thomas
      – PersonEntity:
          Name:
            NameFull: Babin, Danilo
      – PersonEntity:
          Name:
            NameFull: Chen, Min
      – PersonEntity:
          Name:
            NameFull: Callens, Martijn
      – PersonEntity:
          Name:
            NameFull: Brack, Denim
      – PersonEntity:
          Name:
            NameFull: Maes, Helena
      – PersonEntity:
          Name:
            NameFull: Lievens, Jan
      – PersonEntity:
          Name:
            NameFull: Lammens, Marie
      – PersonEntity:
          Name:
            NameFull: Van Sumere, Maxime
      – PersonEntity:
          Name:
            NameFull: Morbée, Lieve
      – PersonEntity:
          Name:
            NameFull: Hautekeete, Simon
      – PersonEntity:
          Name:
            NameFull: Schatteman, Stijn
      – PersonEntity:
          Name:
            NameFull: Jacobs, Tom
      – PersonEntity:
          Name:
            NameFull: Thooft, Willem-Jan
      – PersonEntity:
          Name:
            NameFull: Herregods, Nele
      – PersonEntity:
          Name:
            NameFull: Huysse, Wouter
      – PersonEntity:
          Name:
            NameFull: Jaremko, Jacob L.
      – PersonEntity:
          Name:
            NameFull: Lambert, Robert
      – PersonEntity:
          Name:
            NameFull: Maksymowych, Walter
      – PersonEntity:
          Name:
            NameFull: Laloo, Frederiek
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 11
              Text: Nov2023
              Type: published
              Y: 2023
          Identifiers:
            – Type: issn-print
              Value: 09387994
          Numbering:
            – Type: volume
              Value: 33
            – Type: issue
              Value: 11
          Titles:
            – TitleFull: European Radiology
              Type: main
ResultId 1