View in EDS HTML Full Text PDF Full Text

AI-powered prostate cancer detection: a multi-centre, multi-scanner validation study.

Saved in:
Bibliographic Details
Title: AI-powered prostate cancer detection: a multi-centre, multi-scanner validation study.
Authors: Giganti, Francesco1,2 (AUTHOR) f.giganti@ucl.ac.uk, Moreira da Silva, Nadia3 (AUTHOR), Yeung, Michael3 (AUTHOR), Davies, Lucy3 (AUTHOR), Frary, Amy3 (AUTHOR), Ferrer Rodriguez, Mirjana3 (AUTHOR), Sushentsev, Nikita4 (AUTHOR), Ashley, Nicholas3,5 (AUTHOR), Andreou, Adrian6 (AUTHOR), Bradley, Alison5 (AUTHOR), Wilson, Chris3 (AUTHOR), Maskell, Giles5 (AUTHOR), Brembilla, Giorgio7 (AUTHOR), Caglic, Iztok4 (AUTHOR), Suchánek, Jakub8 (AUTHOR), Budd, Jobie3 (AUTHOR), Arya, Zobair3 (AUTHOR), Aning, Jonathan9 (AUTHOR), Hayes, John10,11 (AUTHOR), De Bono, Mark12 (AUTHOR)
Source: European Radiology. Aug2025, Vol. 35 Issue 8, p4915-4924. 10p.
Subjects: Prostate cancer, Artificial intelligence, Deep learning, Calibration, Detection algorithms, Multihospital systems, Radiologists, Magnetic resonance imaging
Geographic Terms: United Kingdom
Abstract: Objectives: Multi-centre, multi-vendor validation of artificial intelligence (AI) software to detect clinically significant prostate cancer (PCa) using multiparametric magnetic resonance imaging (MRI) is lacking. We compared a new AI solution, validated on a separate dataset from different UK hospitals, to the original multidisciplinary team (MDT)-supported radiologist's interpretations. Materials and methods: A Conformité Européenne (CE)-marked deep-learning (DL) computer-aided detection (CAD) medical device (Pi) was trained to detect Gleason Grade Group (GG) ≥ 2 cancer using retrospective data from the PROSTATEx dataset and five UK hospitals (793 patients). Our separate validation dataset was on six machines from two manufacturers across six sites (252 patients). Data included in the study were from MRI scans performed between August 2018 to October 2022. Patients with a negative MRI who did not undergo biopsy were assumed to be negative (90.4% had prostate-specific antigen density < 0.15 ng/mL2). ROC analysis was used to compare radiologists who used a 5-category suspicion score. Results: GG ≥ 2 prevalence in the validation set was 31%. Evaluated per patient, Pi was non-inferior to radiologists (considering a 10% performance difference as acceptable), with an area under the curve (AUC) of 0.91 vs. 0.95. At the predetermined risk threshold of 3.5, the AI software's sensitivity was 95% and specificity 67%, while radiologists at Prostate Imaging-Reporting and Data Systems/Likert ≥ 3 identified GG ≥ 2 with a sensitivity of 99% and specificity of 73%. AI performed well per-site (AUC ≥ 0.83) at the patient-level independent of scanner age and field strength. Conclusion: Real-world data testing suggests that Pi matches the performance of MDT-supported radiologists in GG ≥ 2 PCa detection and generalises to multiple sites, scanner vendors, and models. Key Points: QuestionThe performance of artificial intelligence-based medical tools for prostate MRI has yet to be evaluated on multi-centre, multi-vendor data to assess generalisability. FindingsA dedicated AI medical tool matches the performance of multidisciplinary team-supported radiologists in prostate cancer detection and generalises to multiple sites and scanners. Clinical relevanceThis software has the potential to support the MRI process for biopsy decision-making and target identification, but future prospective studies, where lesions identified by artificial intelligence are biopsied separately, are needed. [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: 186373218
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: AI-powered prostate cancer detection: a multi-centre, multi-scanner validation study.
– Name: Author
  Label: Authors
  Group: Au
  Data: &lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Giganti%2C+Francesco%22&quot;&gt;Giganti, Francesco&lt;/searchLink&gt;&lt;relatesTo&gt;1,2&lt;/relatesTo&gt; (AUTHOR)&lt;i&gt; f.giganti@ucl.ac.uk&lt;/i&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Moreira+da+Silva%2C+Nadia%22&quot;&gt;Moreira da Silva, Nadia&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Yeung%2C+Michael%22&quot;&gt;Yeung, Michael&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Davies%2C+Lucy%22&quot;&gt;Davies, Lucy&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Frary%2C+Amy%22&quot;&gt;Frary, Amy&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Ferrer+Rodriguez%2C+Mirjana%22&quot;&gt;Ferrer Rodriguez, Mirjana&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Sushentsev%2C+Nikita%22&quot;&gt;Sushentsev, Nikita&lt;/searchLink&gt;&lt;relatesTo&gt;4&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Ashley%2C+Nicholas%22&quot;&gt;Ashley, Nicholas&lt;/searchLink&gt;&lt;relatesTo&gt;3,5&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Andreou%2C+Adrian%22&quot;&gt;Andreou, Adrian&lt;/searchLink&gt;&lt;relatesTo&gt;6&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Bradley%2C+Alison%22&quot;&gt;Bradley, Alison&lt;/searchLink&gt;&lt;relatesTo&gt;5&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Wilson%2C+Chris%22&quot;&gt;Wilson, Chris&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Maskell%2C+Giles%22&quot;&gt;Maskell, Giles&lt;/searchLink&gt;&lt;relatesTo&gt;5&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Brembilla%2C+Giorgio%22&quot;&gt;Brembilla, Giorgio&lt;/searchLink&gt;&lt;relatesTo&gt;7&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Caglic%2C+Iztok%22&quot;&gt;Caglic, Iztok&lt;/searchLink&gt;&lt;relatesTo&gt;4&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Such&#225;nek%2C+Jakub%22&quot;&gt;Such&#225;nek, Jakub&lt;/searchLink&gt;&lt;relatesTo&gt;8&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Budd%2C+Jobie%22&quot;&gt;Budd, Jobie&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Arya%2C+Zobair%22&quot;&gt;Arya, Zobair&lt;/searchLink&gt;&lt;relatesTo&gt;3&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Aning%2C+Jonathan%22&quot;&gt;Aning, Jonathan&lt;/searchLink&gt;&lt;relatesTo&gt;9&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Hayes%2C+John%22&quot;&gt;Hayes, John&lt;/searchLink&gt;&lt;relatesTo&gt;10,11&lt;/relatesTo&gt; (AUTHOR)&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22De+Bono%2C+Mark%22&quot;&gt;De Bono, Mark&lt;/searchLink&gt;&lt;relatesTo&gt;12&lt;/relatesTo&gt; (AUTHOR)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: &lt;searchLink fieldCode=&quot;JN&quot; term=&quot;%22European+Radiology%22&quot;&gt;European Radiology&lt;/searchLink&gt;. Aug2025, Vol. 35 Issue 8, p4915-4924. 10p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: &lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Prostate+cancer%22&quot;&gt;Prostate cancer&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Artificial+intelligence%22&quot;&gt;Artificial intelligence&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Deep+learning%22&quot;&gt;Deep learning&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Calibration%22&quot;&gt;Calibration&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Detection+algorithms%22&quot;&gt;Detection algorithms&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Multihospital+systems%22&quot;&gt;Multihospital systems&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Radiologists%22&quot;&gt;Radiologists&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Magnetic+resonance+imaging%22&quot;&gt;Magnetic resonance imaging&lt;/searchLink&gt;
– Name: SubjectGeographic
  Label: Geographic Terms
  Group: Su
  Data: &lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22United+Kingdom%22&quot;&gt;United Kingdom&lt;/searchLink&gt;
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Objectives: Multi-centre, multi-vendor validation of artificial intelligence (AI) software to detect clinically significant prostate cancer (PCa) using multiparametric magnetic resonance imaging (MRI) is lacking. We compared a new AI solution, validated on a separate dataset from different UK hospitals, to the original multidisciplinary team (MDT)-supported radiologist&#39;s interpretations. Materials and methods: A Conformit&#233; Europ&#233;enne (CE)-marked deep-learning (DL) computer-aided detection (CAD) medical device (Pi) was trained to detect Gleason Grade Group (GG) ≥ 2 cancer using retrospective data from the PROSTATEx dataset and five UK hospitals (793 patients). Our separate validation dataset was on six machines from two manufacturers across six sites (252 patients). Data included in the study were from MRI scans performed between August 2018 to October 2022. Patients with a negative MRI who did not undergo biopsy were assumed to be negative (90.4% had prostate-specific antigen density &lt; 0.15 ng/mL2). ROC analysis was used to compare radiologists who used a 5-category suspicion score. Results: GG ≥ 2 prevalence in the validation set was 31%. Evaluated per patient, Pi was non-inferior to radiologists (considering a 10% performance difference as acceptable), with an area under the curve (AUC) of 0.91 vs. 0.95. At the predetermined risk threshold of 3.5, the AI software&#39;s sensitivity was 95% and specificity 67%, while radiologists at Prostate Imaging-Reporting and Data Systems/Likert ≥ 3 identified GG ≥ 2 with a sensitivity of 99% and specificity of 73%. AI performed well per-site (AUC ≥ 0.83) at the patient-level independent of scanner age and field strength. Conclusion: Real-world data testing suggests that Pi matches the performance of MDT-supported radiologists in GG ≥ 2 PCa detection and generalises to multiple sites, scanner vendors, and models. Key Points: QuestionThe performance of artificial intelligence-based medical tools for prostate MRI has yet to be evaluated on multi-centre, multi-vendor data to assess generalisability. FindingsA dedicated AI medical tool matches the performance of multidisciplinary team-supported radiologists in prostate cancer detection and generalises to multiple sites and scanners. Clinical relevanceThis software has the potential to support the MRI process for biopsy decision-making and target identification, but future prospective studies, where lesions identified by artificial intelligence are biopsied separately, are needed. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: &lt;i&gt;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&#39;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.&lt;/i&gt; (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=186373218
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s00330-024-11323-0
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 10
        StartPage: 4915
    Subjects:
      – SubjectFull: Prostate cancer
        Type: general
      – SubjectFull: Artificial intelligence
        Type: general
      – SubjectFull: Deep learning
        Type: general
      – SubjectFull: Calibration
        Type: general
      – SubjectFull: Detection algorithms
        Type: general
      – SubjectFull: Multihospital systems
        Type: general
      – SubjectFull: Radiologists
        Type: general
      – SubjectFull: Magnetic resonance imaging
        Type: general
      – SubjectFull: United Kingdom
        Type: general
    Titles:
      – TitleFull: AI-powered prostate cancer detection: a multi-centre, multi-scanner validation study.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Giganti, Francesco
      – PersonEntity:
          Name:
            NameFull: Moreira da Silva, Nadia
      – PersonEntity:
          Name:
            NameFull: Yeung, Michael
      – PersonEntity:
          Name:
            NameFull: Davies, Lucy
      – PersonEntity:
          Name:
            NameFull: Frary, Amy
      – PersonEntity:
          Name:
            NameFull: Ferrer Rodriguez, Mirjana
      – PersonEntity:
          Name:
            NameFull: Sushentsev, Nikita
      – PersonEntity:
          Name:
            NameFull: Ashley, Nicholas
      – PersonEntity:
          Name:
            NameFull: Andreou, Adrian
      – PersonEntity:
          Name:
            NameFull: Bradley, Alison
      – PersonEntity:
          Name:
            NameFull: Wilson, Chris
      – PersonEntity:
          Name:
            NameFull: Maskell, Giles
      – PersonEntity:
          Name:
            NameFull: Brembilla, Giorgio
      – PersonEntity:
          Name:
            NameFull: Caglic, Iztok
      – PersonEntity:
          Name:
            NameFull: Suchánek, Jakub
      – PersonEntity:
          Name:
            NameFull: Budd, Jobie
      – PersonEntity:
          Name:
            NameFull: Arya, Zobair
      – PersonEntity:
          Name:
            NameFull: Aning, Jonathan
      – PersonEntity:
          Name:
            NameFull: Hayes, John
      – PersonEntity:
          Name:
            NameFull: De Bono, Mark
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 08
              Text: Aug2025
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 09387994
          Numbering:
            – Type: volume
              Value: 35
            – Type: issue
              Value: 8
          Titles:
            – TitleFull: European Radiology
              Type: main
ResultId 1