{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,16]],"date-time":"2026-09-16T01:05:16Z","timestamp":1789520716050,"version":"build-2803163510"},"reference-count":28,"publisher":"Springer Science and Business Media LLC","issue":"6","license":[{"start":{"date-parts":[[2026,1,13]],"date-time":"2026-01-13T00:00:00Z","timestamp":1768262400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,1,13]],"date-time":"2026-01-13T00:00:00Z","timestamp":1768262400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100008730","name":"Kreftforeningen","doi-asserted-by":"publisher","award":["214931"],"award-info":[{"award-number":["214931"]}],"id":[{"id":"10.13039\/100008730","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Eur Radiol"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objectives<\/jats:title>\n                    <jats:p>Studies have reported promising results regarding artificial intelligence (AI) as a tool for improved mammographic screening interpretive performance. We analyzed AI malignancy risk scores from two versions of the same commercial AI model.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and methods<\/jats:title>\n                    <jats:p>This retrospective cohort study used data from 117,709 screening examinations performed in BreastScreen Norway 2009\u20132018. The mammograms were processed by two versions of the commercially available AI model, Transpara (version 1.7 and 2.1). The distributions of exam-level risk scores (AI score 1\u201310) and risk categories were evaluated for both AI versions on all examinations, including 737 screen-detected and 200 interval cancers. Scores between 1\u20137 were categorized as low risk, 8\u20139 as intermediate risk, and 10 as high risk of malignancy.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Area under the receiver operating curve was 0.908 (95% CI: 0.986\u20130.920) for version 1.7 and 0.928 (95% CI: 0.917\u20130.939) for 2.1 when screen-detected and interval cancers were considered as positive cases (\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.001). A total of 87.1% (642\/737) and 93.5% (689\/737) of the screen-detected cancers had an AI score of 10 with version 1.7 and 2.1, respectively. Among interval cancers, 45.0% (90\/200) had AI score 10 with version 1.7 and 44.5% (89\/200) had AI score 10 with version 2.1.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>A higher proportion of screen-detected breast cancers had the highest AI score of 10 with the newer version of the AI model compared to the older version. For interval cancers, there was no difference in the proportion of cases assigned to the highest score between the two versions.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Key Points<\/jats:title>\n                    <jats:p>\n                      <jats:bold>\n                        <jats:italic>Question<\/jats:italic>\n                      <\/jats:bold>\n                      <jats:italic>Studies have reported promising results regarding the use of AI in mammography screening, but comparisons of updated versus older versions are less studied<\/jats:italic>\n                      .\n                    <\/jats:p>\n                    <jats:p>\n                      <jats:bold>\n                        <jats:italic>Findings<\/jats:italic>\n                      <\/jats:bold>\n                      <jats:italic>In our study, 87.1% (642\/737) of the screen-detected cancers were classified with a high malignancy risk score by the old version, while it was 93.5% (689\/737) for the newer version<\/jats:italic>\n                      .\n                    <\/jats:p>\n                    <jats:p>\n                      <jats:bold>\n                        <jats:italic>Clinical relevance<\/jats:italic>\n                      <\/jats:bold>\n                      <jats:italic>Understanding how version updates of AI models might impact screening mammography performance will be important for future quality assurance and validation of AI models<\/jats:italic>\n                      .\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.1007\/s00330-025-12240-6","type":"journal-article","created":{"date-parts":[[2026,1,13]],"date-time":"2026-01-13T10:34:55Z","timestamp":1768300495000},"page":"4422-4433","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Performance across different versions of an artificial intelligence model for screen-reading of mammograms"],"prefix":"10.1007","volume":"36","author":[{"given":"Marthe","family":"Larsen","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christoph I.","family":"Lee","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marie B.","family":"Bergan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"\u00c5sne S.","family":"Holen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"H\u00e5kon","family":"Lund-Hanssen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Solveig R.","family":"Hoff","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Steinar","family":"Auensen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jan F.","family":"Nyg\u00e5rd","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kristina","family":"L\u00e5ng","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yan","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Giske","family":"Ursin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0178-8939","authenticated-orcid":false,"given":"Solveig","family":"Hofvind","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,1,13]]},"reference":[{"key":"12240_CR1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.223338","volume":"306","author":"CK Kuhl","year":"2023","unstructured":"Kuhl CK (2023) What the future holds for the screening, diagnosis, and treatment of breast cancer. Radiology 306:e223338","journal-title":"Radiology"},{"key":"12240_CR2","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.230018","volume":"306","author":"L Moy","year":"2023","unstructured":"Moy L (2023) Change is good: the evolution and future of breast imaging. Radiology 306:e230018","journal-title":"Radiology"},{"key":"12240_CR3","doi-asserted-by":"publisher","first-page":"n1872","DOI":"10.1136\/bmj.n1872","volume":"374","author":"K Freeman","year":"2021","unstructured":"Freeman K, Geppert J, Stinton C et al (2021) Use of artificial intelligence for image analysis in breast cancer screening programmes: systematic review of test accuracy. BMJ 374:n1872","journal-title":"BMJ"},{"key":"12240_CR4","doi-asserted-by":"publisher","first-page":"e230375","DOI":"10.1148\/ryai.230375","volume":"6","author":"M Larsen","year":"2024","unstructured":"Larsen M, Olstad CF, Lee CI et al (2024) Performance of an artificial intelligence system for breast cancer detection on screening mammograms from BreastScreen Norway. Radiol Artif Intell 6:e230375","journal-title":"Radiol Artif Intell"},{"key":"12240_CR5","doi-asserted-by":"publisher","first-page":"1687","DOI":"10.1007\/s00330-020-07165-1","volume":"31","author":"K L\u00e5ng","year":"2021","unstructured":"L\u00e5ng K, Dustler M, Dahlblom V, \u00c5kesson A, Andersson I, Zackrisson S (2021) Identifying normal mammograms in a large screening population using artificial intelligence. Eur Radiol 31:1687\u20131692","journal-title":"Eur Radiol"},{"key":"12240_CR6","doi-asserted-by":"publisher","first-page":"e703","DOI":"10.1016\/S2589-7500(23)00153-X","volume":"5","author":"K Dembrower","year":"2023","unstructured":"Dembrower K, Crippa A, Col\u00f3n E, Eklund M, Strand F (2023) Artificial intelligence for breast cancer detection in screening mammography in Sweden: a prospective, population-based, paired-reader, non-inferiority study. Lancet Digit Health 5:e703\u2013e711","journal-title":"Lancet Digit Health"},{"key":"12240_CR7","doi-asserted-by":"publisher","first-page":"936","DOI":"10.1016\/S1470-2045(23)00298-X","volume":"24","author":"K L\u00e5ng","year":"2023","unstructured":"L\u00e5ng K, Josefsson V, Larsson AM et al (2023) Artificial intelligence-supported screen reading versus standard double reading in the Mammography Screening with Artificial Intelligence trial (MASAI): a clinical safety analysis of a randomised, controlled, non-inferiority, single-blinded, screening accuracy study. Lancet Oncol 24:936\u2013944","journal-title":"Lancet Oncol"},{"key":"12240_CR8","doi-asserted-by":"publisher","unstructured":"Eisemann N, Bunk S, Mukama T et al (2025) Nationwide real-world implementation of AI for cancer detection in population-based mammography screening. Nat Med. https:\/\/doi.org\/10.1038\/s41591-024-03408-6","DOI":"10.1038\/s41591-024-03408-6"},{"key":"12240_CR9","doi-asserted-by":"publisher","first-page":"e175","DOI":"10.1016\/S2589-7500(24)00267-X","volume":"7","author":"V Hernstr\u00f6m","year":"2025","unstructured":"Hernstr\u00f6m V, Josefsson V, Sartor H et al (2025) Screening performance and characteristics of breast cancer detected in the Mammography Screening with Artificial Intelligence trial (MASAI): a randomised, controlled, parallel-group, non-inferiority, single-blinded, screening accuracy study. Lancet Digit Health 7:e175\u2013e183","journal-title":"Lancet Digit Health"},{"key":"12240_CR10","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-025-57469-3","volume":"16","author":"Y-W Chang","year":"2025","unstructured":"Chang Y-W, Ryu JK, An JK et al (2025) Artificial intelligence for breast cancer screening in mammography (AI-STREAM): preliminary analysis of a prospective multicenter cohort study. Nat Commun 16:2248","journal-title":"Nat Commun"},{"key":"12240_CR11","doi-asserted-by":"publisher","first-page":"3958","DOI":"10.1007\/s00330-023-10426-4","volume":"34","author":"E El\u00edas-Cabot","year":"2024","unstructured":"El\u00edas-Cabot E, Romero-Mart\u00edn S, Raya-Povedano JL, Brehl AK, \u00c1lvarez-Benito M (2024) Impact of real-life use of artificial intelligence as support for human reading in a population-based breast cancer screening program with mammography and tomosynthesis. Eur Radiol 34:3958\u20133966","journal-title":"Eur Radiol"},{"key":"12240_CR12","doi-asserted-by":"publisher","unstructured":"Hovda T, Larsen M, Bergan MB, Gjesvik J, Akslen LA, Hofvind S (2025) Retrospective evaluation of a CE-marked AI system, including 1,017,208 mammography screening examinations. Eur Radiol. https:\/\/doi.org\/10.1007\/s00330-025-11521-4","DOI":"10.1007\/s00330-025-11521-4"},{"key":"12240_CR13","doi-asserted-by":"publisher","first-page":"112100","DOI":"10.1016\/j.ejrad.2025.112100","volume":"187","author":"AG Taib","year":"2025","unstructured":"Taib AG, James JJ, Partridge GJW, Chen Y (2025) Keeping AI on track: regular monitoring of algorithmic updates in mammography. Eur J Radiol 187:112100","journal-title":"Eur J Radiol"},{"key":"12240_CR14","unstructured":"Lovdata (2015) Lov om helseregistre og behandling av helseopplysninger (helseregisterloven). Available via https:\/\/lovdata.no\/dokument\/NL\/lov\/2014-06-20-43. Accessed 4 Apr 2023"},{"key":"12240_CR15","unstructured":"Lovdata (2001) Kreftregisterforskriften. Available via https:\/\/lovdata.no\/dokument\/SF\/forskrift\/2001-12-21-1477. Accessed 30 Nov 2024"},{"key":"12240_CR16","unstructured":"Bj\u00f8rnson E, Holen \u00c5S, Sagstad S et al (2022) BreastScreen Norway: 25 years of organized screening. https:\/\/www.fhi.no\/globalassets\/mammografiprogrammet\/rapporter-og-publikasjoner\/2022-25-arsrapport_webversjon.pdf"},{"key":"12240_CR17","unstructured":"Hofvind S, Tsuruda K, Mangerud G, Ertzaas A (2017) The Norwegian breast cancer screening program, 1996\u20132016: celebrating 20 years of organised mammographic screening. In: Cancer in Norway 2016\u2014cancer incidence, mortality, survival and prevalence in Norway. https:\/\/www.fhi.no\/globalassets\/cancer-in-norway\/2016\/mammo_cin2016_special_issue_web.pdf"},{"key":"12240_CR18","doi-asserted-by":"publisher","first-page":"502","DOI":"10.1148\/radiol.212381","volume":"303","author":"M Larsen","year":"2022","unstructured":"Larsen M, Aglen CF, Lee CI et al (2022) Artificial intelligence evaluation of 122\u2009969 mammography examinations from a population-based screening program. Radiology 303:502\u2013511","journal-title":"Radiology"},{"key":"12240_CR19","doi-asserted-by":"publisher","first-page":"e230989","DOI":"10.1148\/radiol.230989","volume":"309","author":"M Larsen","year":"2023","unstructured":"Larsen M, Olstad CF, Koch HW et al (2023) AI risk score on screening mammograms preceding breast cancer diagnosis. Radiology 309:e230989","journal-title":"Radiology"},{"key":"12240_CR20","unstructured":"U.S. Food & Drug Administration (2025) 510(k) Premarket Notification. Available via https:\/\/www.accessdata.fda.gov\/scripts\/cdrh\/cfdocs\/cfpmn\/pmn.cfm?ID=K241831. Accessed 14 Apr 2024"},{"key":"12240_CR21","unstructured":"U.S. Food & Drug Administration (2021) 510(k) Premarket Notification. Available via https:\/\/www.accessdata.fda.gov\/scripts\/cdrh\/cfdocs\/cfpmn\/pmn.cfm?ID=K210404. Accessed 14 Apr 2024"},{"key":"12240_CR22","doi-asserted-by":"publisher","first-page":"2206","DOI":"10.1093\/annonc\/mdt303","volume":"24","author":"A Goldhirsch","year":"2013","unstructured":"Goldhirsch A, Winer EP, Coates AS et al (2013) Personalizing the treatment of women with early breast cancer: highlights of the St Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2013. Ann Oncol 24:2206\u20132223","journal-title":"Ann Oncol"},{"key":"12240_CR23","unstructured":"Sickles E, D\u2019Orsi C, Bassett L et al (2013) ACR BI-RADS\u00ae mammography. In: ACR BI-RADS\u00ae Atlas, Breast Imaging Reporting and Data System, 5th edn. American College of Radiology, Reston"},{"key":"12240_CR24","unstructured":"StataCorp. (2013) roccomp\u2014tests of equality of ROC areas. Available via https:\/\/www.stata.com\/manuals13\/rroccomp.pdf. Accessed 18 Sep 2025"},{"key":"12240_CR25","doi-asserted-by":"publisher","unstructured":"Martiniussen MA, Larsen M, Hovda T et al (2025) Performance of two deep learning-based AI models for breast cancer detection and localization on screening mammograms from BreastScreen Norway. Radiol Artif Intell. https:\/\/doi.org\/10.1148\/ryai.240039:e240039","DOI":"10.1148\/ryai.240039:e240039"},{"key":"12240_CR26","doi-asserted-by":"publisher","unstructured":"Hovda T, Hoff SR, Larsen M, Romundstad L, Sahlberg KK, Hofvind S (2021) True and missed interval cancer in organized mammographic screening: a retrospective review study of diagnostic and prior screening mammograms. Acad Radiol. https:\/\/doi.org\/10.1016\/j.acra.2021.03.022","DOI":"10.1016\/j.acra.2021.03.022"},{"key":"12240_CR27","doi-asserted-by":"publisher","DOI":"10.1038\/s41523-017-0014-x","volume":"3","author":"N Houssami","year":"2017","unstructured":"Houssami N, Hunter K (2017) The epidemiology, radiology and biological characteristics of interval breast cancers in population mammography screening. NPJ Breast Cancer 3:12","journal-title":"NPJ Breast Cancer"},{"key":"12240_CR28","doi-asserted-by":"publisher","first-page":"6158","DOI":"10.1007\/s00330-024-10662-2","volume":"34","author":"HW Koch","year":"2024","unstructured":"Koch HW, Larsen M, Bartsch H et al (2024) How do AI markings on screening mammograms correspond to cancer location? An informed review of 270 breast cancer cases in BreastScreen Norway. Eur Radiol 34:6158\u20136167","journal-title":"Eur Radiol"}],"container-title":["European Radiology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00330-025-12240-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s00330-025-12240-6","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s00330-025-12240-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,26]],"date-time":"2026-05-26T12:26:31Z","timestamp":1779798391000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s00330-025-12240-6"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,13]]},"references-count":28,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["12240"],"URL":"https:\/\/doi.org\/10.1007\/s00330-025-12240-6","relation":{},"ISSN":["1432-1084"],"issn-type":[{"value":"1432-1084","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,13]]},"assertion":[{"value":"4 June 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"18 September 2025","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"1 December 2025","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"13 January 2026","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Compliance with ethical standards"}},{"value":"The scientific guarantor of this publication is Solveig Hofvind.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Guarantor"}},{"value":"Dr. C. Lee reports textbook royalties from McGraw-Hill, Inc., Oxford University Press, and UpToDate, Inc., research consulting fees from DeepHealth\/RadNet, and personal fees for journal editorial board work from the American College of Radiology, all outside the submitted work. K.L. is affiliated with Unilabs. The other authors declare no conflicts of interest.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}},{"value":"No complex statistical methods were necessary for this paper, but Marthe Larsen is a statistician.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Statistics and biometry"}},{"value":"Written informed consent was waived by the Institutional Review Board. The data was disclosed with a legal basis in the Cancer Registry Regulations section 3\u20131 and the Personal Health Filing System Act section 19a to 19h.","order":5,"name":"Ethics","group":{"name":"EthicsHeading","label":"Informed consent"}},{"value":"Institutional Review Board approval was obtained, i.e., the study was approved by the Regional Committee for Medical and Health Research Ethics (2018\/13294).","order":6,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}},{"value":"Overall performance and possible strategies for combining AI and radiologists for almost an identical sample of the screening examinations (122,969 examinations) have previously been reported for version 1.7 (\n                      \n                      ). The sample with results from 1.7 has also been included in an analysis of AI score on mammograms preceding a cancer diagnosis (\n                      \n                      ). Since we could merge results from version 1.7 and 2.1 at an individual examination level, we reported the overall performance for 1.7 in this paper as well, since the sample included about 5000 fewer examinations.","order":7,"name":"Ethics","group":{"name":"EthicsHeading","label":"Study subjects or cohorts overlap"}},{"order":8,"name":"Ethics","group":{"name":"EthicsHeading","label":"Methodology"}}]}}