View in EDS HTML Full Text PDF Full Text

Can we reduce the workload of mammographic screening by automatic identification of normal exams with artificial intelligence? A feasibility study.

Saved in:
Bibliographic Details
Title: Can we reduce the workload of mammographic screening by automatic identification of normal exams with artificial intelligence? A feasibility study.
Authors: Rodriguez-Ruiz, Alejandro1,2, Lång, Kristina3, Gubern-Merida, Albert2, Teuwen, Jonas1, Broeders, Mireille4,5, Gennaro, Gisella6, Clauser, Paola7, Helbich, Thomas H.7, Chevalier, Margarita8, Mertelmeier, Thomas9, Wallis, Matthew G.10, Andersson, Ingvar11, Zackrisson, Sophia12, Sechopoulos, Ioannis1,5, Mann, Ritse M.1 Ritse.Mann@radboudumc.nl
Source: European Radiology. Sep2019, Vol. 29 Issue 9, p4825-4832. 8p. 1 Black and White Photograph, 3 Graphs.
Subjects: Artificial intelligence, Automatic identification, Receiver operating characteristic curves, Early detection of cancer, Breast tumors, Probability theory, Pilot projects, Diagnostic errors, Mammograms, Medical screening, Employees' workload
Abstract: Purpose: To study the feasibility of automatically identifying normal digital mammography (DM) exams with artificial intelligence (AI) to reduce the breast cancer screening reading workload.Methods and Materials: A total of 2652 DM exams (653 cancer) and interpretations by 101 radiologists were gathered from nine previously performed multi-reader multi-case receiver operating characteristic (MRMC ROC) studies. An AI system was used to obtain a score between 1 and 10 for each exam, representing the likelihood of cancer present. Using all AI scores between 1 and 9 as possible thresholds, the exams were divided into groups of low- and high likelihood of cancer present. It was assumed that, under the pre-selection scenario, only the high-likelihood group would be read by radiologists, while all low-likelihood exams would be reported as normal. The area under the reader-averaged ROC curve (AUC) was calculated for the original evaluations and for the pre-selection scenarios and compared using a non-inferiority hypothesis.Results: Setting the low/high-likelihood threshold at an AI score of 5 (high likelihood > 5) results in a trade-off of approximately halving (- 47%) the workload to be read by radiologists while excluding 7% of true-positive exams. Using an AI score of 2 as threshold yields a workload reduction of 17% while only excluding 1% of true-positive exams. Pre-selection did not change the average AUC of radiologists (inferior 95% CI > - 0.05) for any threshold except at the extreme AI score of 9.Conclusion: It is possible to automatically pre-select exams using AI to significantly reduce the breast cancer screening reading workload.Key Points: • There is potential to use artificial intelligence to automatically reduce the breast cancer screening reading workload by excluding exams with a low likelihood of cancer. • The exclusion of exams with the lowest likelihood of cancer in screening might not change radiologists' breast cancer detection performance. • When excluding exams with the lowest likelihood of cancer, the decrease in true-positive recalls would be balanced by a simultaneous reduction in false-positive recalls. [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: 137908435
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Can we reduce the workload of mammographic screening by automatic identification of normal exams with artificial intelligence? A feasibility study.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Rodriguez-Ruiz%2C+Alejandro%22">Rodriguez-Ruiz, Alejandro</searchLink><relatesTo>1,2</relatesTo><br /><searchLink fieldCode="AR" term="%22Lång%2C+Kristina%22">Lång, Kristina</searchLink><relatesTo>3</relatesTo><br /><searchLink fieldCode="AR" term="%22Gubern-Merida%2C+Albert%22">Gubern-Merida, Albert</searchLink><relatesTo>2</relatesTo><br /><searchLink fieldCode="AR" term="%22Teuwen%2C+Jonas%22">Teuwen, Jonas</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22Broeders%2C+Mireille%22">Broeders, Mireille</searchLink><relatesTo>4,5</relatesTo><br /><searchLink fieldCode="AR" term="%22Gennaro%2C+Gisella%22">Gennaro, Gisella</searchLink><relatesTo>6</relatesTo><br /><searchLink fieldCode="AR" term="%22Clauser%2C+Paola%22">Clauser, Paola</searchLink><relatesTo>7</relatesTo><br /><searchLink fieldCode="AR" term="%22Helbich%2C+Thomas+H%2E%22">Helbich, Thomas H.</searchLink><relatesTo>7</relatesTo><br /><searchLink fieldCode="AR" term="%22Chevalier%2C+Margarita%22">Chevalier, Margarita</searchLink><relatesTo>8</relatesTo><br /><searchLink fieldCode="AR" term="%22Mertelmeier%2C+Thomas%22">Mertelmeier, Thomas</searchLink><relatesTo>9</relatesTo><br /><searchLink fieldCode="AR" term="%22Wallis%2C+Matthew+G%2E%22">Wallis, Matthew G.</searchLink><relatesTo>10</relatesTo><br /><searchLink fieldCode="AR" term="%22Andersson%2C+Ingvar%22">Andersson, Ingvar</searchLink><relatesTo>11</relatesTo><br /><searchLink fieldCode="AR" term="%22Zackrisson%2C+Sophia%22">Zackrisson, Sophia</searchLink><relatesTo>12</relatesTo><br /><searchLink fieldCode="AR" term="%22Sechopoulos%2C+Ioannis%22">Sechopoulos, Ioannis</searchLink><relatesTo>1,5</relatesTo><br /><searchLink fieldCode="AR" term="%22Mann%2C+Ritse+M%2E%22">Mann, Ritse M.</searchLink><relatesTo>1</relatesTo><i> Ritse.Mann@radboudumc.nl</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22European+Radiology%22">European Radiology</searchLink>. Sep2019, Vol. 29 Issue 9, p4825-4832. 8p. 1 Black and White Photograph, 3 Graphs.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Artificial+intelligence%22">Artificial intelligence</searchLink><br /><searchLink fieldCode="DE" term="%22Automatic+identification%22">Automatic identification</searchLink><br /><searchLink fieldCode="DE" term="%22Receiver+operating+characteristic+curves%22">Receiver operating characteristic curves</searchLink><br /><searchLink fieldCode="DE" term="%22Early+detection+of+cancer%22">Early detection of cancer</searchLink><br /><searchLink fieldCode="DE" term="%22Breast+tumors%22">Breast tumors</searchLink><br /><searchLink fieldCode="DE" term="%22Probability+theory%22">Probability theory</searchLink><br /><searchLink fieldCode="DE" term="%22Pilot+projects%22">Pilot projects</searchLink><br /><searchLink fieldCode="DE" term="%22Diagnostic+errors%22">Diagnostic errors</searchLink><br /><searchLink fieldCode="DE" term="%22Mammograms%22">Mammograms</searchLink><br /><searchLink fieldCode="DE" term="%22Medical+screening%22">Medical screening</searchLink><br /><searchLink fieldCode="DE" term="%22Employees'+workload%22">Employees' workload</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: <bold>Purpose: </bold>To study the feasibility of automatically identifying normal digital mammography (DM) exams with artificial intelligence (AI) to reduce the breast cancer screening reading workload.<bold>Methods and Materials: </bold>A total of 2652 DM exams (653 cancer) and interpretations by 101 radiologists were gathered from nine previously performed multi-reader multi-case receiver operating characteristic (MRMC ROC) studies. An AI system was used to obtain a score between 1 and 10 for each exam, representing the likelihood of cancer present. Using all AI scores between 1 and 9 as possible thresholds, the exams were divided into groups of low- and high likelihood of cancer present. It was assumed that, under the pre-selection scenario, only the high-likelihood group would be read by radiologists, while all low-likelihood exams would be reported as normal. The area under the reader-averaged ROC curve (AUC) was calculated for the original evaluations and for the pre-selection scenarios and compared using a non-inferiority hypothesis.<bold>Results: </bold>Setting the low/high-likelihood threshold at an AI score of 5 (high likelihood > 5) results in a trade-off of approximately halving (- 47%) the workload to be read by radiologists while excluding 7% of true-positive exams. Using an AI score of 2 as threshold yields a workload reduction of 17% while only excluding 1% of true-positive exams. Pre-selection did not change the average AUC of radiologists (inferior 95% CI > - 0.05) for any threshold except at the extreme AI score of 9.<bold>Conclusion: </bold>It is possible to automatically pre-select exams using AI to significantly reduce the breast cancer screening reading workload.<bold>Key Points: </bold>• There is potential to use artificial intelligence to automatically reduce the breast cancer screening reading workload by excluding exams with a low likelihood of cancer. • The exclusion of exams with the lowest likelihood of cancer in screening might not change radiologists' breast cancer detection performance. • When excluding exams with the lowest likelihood of cancer, the decrease in true-positive recalls would be balanced by a simultaneous reduction in false-positive recalls. [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=137908435
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s00330-019-06186-9
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 8
        StartPage: 4825
    Subjects:
      – SubjectFull: Artificial intelligence
        Type: general
      – SubjectFull: Automatic identification
        Type: general
      – SubjectFull: Receiver operating characteristic curves
        Type: general
      – SubjectFull: Early detection of cancer
        Type: general
      – SubjectFull: Breast tumors
        Type: general
      – SubjectFull: Probability theory
        Type: general
      – SubjectFull: Pilot projects
        Type: general
      – SubjectFull: Diagnostic errors
        Type: general
      – SubjectFull: Mammograms
        Type: general
      – SubjectFull: Medical screening
        Type: general
      – SubjectFull: Employees' workload
        Type: general
    Titles:
      – TitleFull: Can we reduce the workload of mammographic screening by automatic identification of normal exams with artificial intelligence? A feasibility study.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Rodriguez-Ruiz, Alejandro
      – PersonEntity:
          Name:
            NameFull: Lång, Kristina
      – PersonEntity:
          Name:
            NameFull: Gubern-Merida, Albert
      – PersonEntity:
          Name:
            NameFull: Teuwen, Jonas
      – PersonEntity:
          Name:
            NameFull: Broeders, Mireille
      – PersonEntity:
          Name:
            NameFull: Gennaro, Gisella
      – PersonEntity:
          Name:
            NameFull: Clauser, Paola
      – PersonEntity:
          Name:
            NameFull: Helbich, Thomas H.
      – PersonEntity:
          Name:
            NameFull: Chevalier, Margarita
      – PersonEntity:
          Name:
            NameFull: Mertelmeier, Thomas
      – PersonEntity:
          Name:
            NameFull: Wallis, Matthew G.
      – PersonEntity:
          Name:
            NameFull: Andersson, Ingvar
      – PersonEntity:
          Name:
            NameFull: Zackrisson, Sophia
      – PersonEntity:
          Name:
            NameFull: Sechopoulos, Ioannis
      – PersonEntity:
          Name:
            NameFull: Mann, Ritse M.
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 09
              Text: Sep2019
              Type: published
              Y: 2019
          Identifiers:
            – Type: issn-print
              Value: 09387994
          Numbering:
            – Type: volume
              Value: 29
            – Type: issue
              Value: 9
          Titles:
            – TitleFull: European Radiology
              Type: main
ResultId 1