{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,12]],"date-time":"2026-11-12T10:20:47Z","timestamp":1794478847293,"version":"4.9.1"},"reference-count":28,"publisher":"Springer Science and Business Media LLC","issue":"3","license":[{"start":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T00:00:00Z","timestamp":1773100800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T00:00:00Z","timestamp":1773100800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"AAC\/NIHR - AI_AWARD02288"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Nat Cancer"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Artificial intelligence (AI) promises to enhance breast cancer screening. Here we evaluated Google\u2019s mammography AI system (version 1.2) across two phases: a retrospective study using 115,973 mammograms from five National Health Service screening services with 39-month follow-up and prospective noninterventional feasibility deployment at 12 sites (9,266 cases). The primary endpoint was AI sensitivity and specificity versus first reader using a 5% noninferiority margin. The secondary endpoints were performance versus second or consensus readers and breast-level analyses. Retrospectively, AI achieved superior sensitivity (0.541 versus 0.437 for first reader,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009&lt;\u20090.001) and noninferior specificity (0.943 versus 0.952,\n                    <jats:italic>P<\/jats:italic>\n                    \u2009&lt;\u20090.001). Cancer detection rate increased from 7.54 to 9.33 per 1,000 women, with AI detecting 25.0% of interval cancers. Performance was particularly strong for first screens (39.3% fewer recalls, 8.8% higher detection) and invasive cancers. No systematic demographic disparities were observed. Simulated second-reader replacement reduced reading time by 32% while increasing detection by 17.7%. Prospective deployment confirmed technical feasibility but revealed a distribution shift requiring threshold recalibration. Implementation requires adaptive calibration and continuous monitoring to ensure safety and equity.\n                  <\/jats:p>","DOI":"10.1038\/s43018-026-01127-0","type":"journal-article","created":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T10:01:45Z","timestamp":1773136905000},"page":"494-506","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":20,"title":["Diagnostic accuracy, fairness and clinical implementation of AI for breast cancer screening: results of multicenter retrospective and prospective technical feasibility studies"],"prefix":"10.1038","volume":"7","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1246-844X","authenticated-orcid":false,"given":"Christopher J.","family":"Kelly","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marc","family":"Wilson","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8296-8917","authenticated-orcid":false,"given":"Lucy M.","family":"Warren","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0063-8082","authenticated-orcid":false,"given":"Richard","family":"Sidebottom","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mark","family":"Halling-Brown","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lin","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Megumi","family":"Morigami","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0547-1226","authenticated-orcid":false,"given":"Jenny","family":"Venton","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4264-8329","authenticated-orcid":false,"given":"Reena","family":"Chopra","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jane","family":"Chang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jonathan","family":"Dixon","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fiona J.","family":"Gilbert","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1114-3184","authenticated-orcid":false,"given":"Daniel I.","family":"Golden","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-1594-3335","authenticated-orcid":false,"given":"Elzbieta","family":"Gruzewska","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lesley","family":"Honeyfield","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Amandeep","family":"Hujan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Delara","family":"Khodabakhshi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Emma","family":"Lewis","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Namrata","family":"Malhotra","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rachita","family":"Mallya","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Della","family":"Ogunleye","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Charlotte","family":"Purdy","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8269-5779","authenticated-orcid":false,"given":"Rory","family":"Sayres","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marcin","family":"Sieniek","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-5506-3859","authenticated-orcid":false,"given":"Tsvetina","family":"Stoycheva","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aminata","family":"Sy","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Susan","family":"Thomas","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Dominic","family":"Ward","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lihong","family":"Xi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shawn","family":"Xu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3783-3172","authenticated-orcid":false,"given":"Shravya","family":"Shetty","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ara","family":"Darzi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3491-257X","authenticated-orcid":false,"given":"Kenneth","family":"Young","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hema","family":"Purushothaman","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lisanne","family":"Khoo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mamatha","family":"Reddy","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1668-0672","authenticated-orcid":false,"given":"Hutan","family":"Ashrafian","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9542-7880","authenticated-orcid":false,"given":"Deborah","family":"Cunningham","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,3,10]]},"reference":[{"key":"1127_CR1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.222639","volume":"307","author":"JH Yoon","year":"2023","unstructured":"Yoon, J. H. et al. Standalone AI for breast cancer detection at screening digital mammography and digital breast tomosynthesis: a systematic review and meta-analysis. Radiology 307, e222639 (2023).","journal-title":"Radiology"},{"key":"1127_CR2","doi-asserted-by":"publisher","first-page":"e232479","DOI":"10.1148\/radiol.232479","volume":"311","author":"AD Lauritzen","year":"2024","unstructured":"Lauritzen, A. D. et al. Early indicators of the impact of using AI in mammography screening for breast cancer. Radiology 311, e232479 (2024).","journal-title":"Radiology"},{"key":"1127_CR3","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. et al. 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 (2023).","journal-title":"Lancet Oncol."},{"key":"1127_CR4","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. et al. 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 (2023).","journal-title":"Lancet Digit. Health"},{"key":"1127_CR5","doi-asserted-by":"publisher","first-page":"e558","DOI":"10.1016\/S2589-7500(22)00088-7","volume":"4","author":"S Taylor-Phillips","year":"2022","unstructured":"Taylor-Phillips, S. et al. UK National Screening Committee\u2019s approach to reviewing evidence on artificial intelligence in breast cancer screening. Lancet Digit. Health 4, e558\u2013e565 (2022).","journal-title":"Lancet Digit. Health"},{"key":"1127_CR6","doi-asserted-by":"publisher","first-page":"2737","DOI":"10.1007\/s00330-016-4612-z","volume":"27","author":"F Sardanelli","year":"2017","unstructured":"Sardanelli, F. et al. Position paper on screening for breast cancer by the European Society of Breast Imaging (EUSOBI) and 30 national breast radiology bodies from Austria, Belgium, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Israel, Lithuania, Moldova, The Netherlands, Norway, Poland, Portugal, Romania, Serbia, Slovakia, Spain, Sweden, Switzerland and Turkey. Eur. Radiol. 27, 2737\u20132743 (2017).","journal-title":"Eur. Radiol."},{"key":"1127_CR7","doi-asserted-by":"publisher","first-page":"57","DOI":"10.4048\/jbc.2022.25.e4","volume":"25","author":"Y-W Chang","year":"2022","unstructured":"Chang, Y.-W. et al. Artificial intelligence for breast cancer screening in mammography (AI-STREAM): a prospective multicenter study design in Korea using AI-based CADe\/x. J. Breast Cancer 25, 57\u201368 (2022).","journal-title":"J. Breast Cancer"},{"key":"1127_CR8","doi-asserted-by":"publisher","first-page":"umae011","DOI":"10.1093\/radadv\/umae011","volume":"1","author":"E Nakai","year":"2024","unstructured":"Nakai, E. et al. Artificial intelligence as a second reader for screening mammography. Radiol. Adv. 1, umae011 (2024).","journal-title":"Radiol. Adv."},{"key":"1127_CR9","doi-asserted-by":"publisher","first-page":"89","DOI":"10.1038\/s41586-019-1799-6","volume":"577","author":"SM McKinney","year":"2020","unstructured":"McKinney, S. M. et al. International evaluation of an AI system for breast cancer screening. Nature 577, 89\u201394 (2020).","journal-title":"Nature"},{"key":"1127_CR10","unstructured":"NHS England. NHS Breast screening programme screening standards valid for data collected from 1 April 2021. GOV.UK https:\/\/www.gov.uk\/government\/publications\/breast-screening-consolidated-programme-standards\/nhs-breast-screening-programme-screening-standards-valid-for-data-collected-from-1-april-2021 (2021)."},{"key":"1127_CR11","doi-asserted-by":"publisher","first-page":"4287","DOI":"10.3390\/cancers13174287","volume":"13","author":"S \u0141ukasiewicz","year":"2021","unstructured":"\u0141ukasiewicz, S. Breast cancer\u2014epidemiology, risk factors, classification, prognostic markers, and current treatment strategies\u2014an updated review. Cancers (Basel) 13, 4287 (2021).","journal-title":"Cancers (Basel)"},{"key":"1127_CR12","doi-asserted-by":"publisher","first-page":"1828","DOI":"10.1001\/jamainternmed.2015.5231","volume":"175","author":"CD Lehman","year":"2015","unstructured":"Lehman, C. D. et al. Diagnostic accuracy of digital screening mammography with and without computer-aided detection. JAMA Intern. Med. 175, 1828\u20131837 (2015).","journal-title":"JAMA Intern. Med."},{"key":"1127_CR13","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.220146","volume":"5","author":"CF de Vries","year":"2023","unstructured":"de Vries, C. F. et al. Impact of different mammography systems on artificial intelligence performance in breast cancer screening. Radiol. Artif. Intell. 5, e220146 (2023).","journal-title":"Radiol. Artif. Intell."},{"key":"1127_CR14","doi-asserted-by":"publisher","unstructured":"Warren, L. M. et al. Impact of using artificial intelligence as a second reader in breast screening including arbitration. Nat. Cancer https:\/\/doi.org\/10.1038\/s43018-026-01128-z (2026).","DOI":"10.1038\/s43018-026-01128-z"},{"key":"1127_CR15","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.2020200103","volume":"3","author":"MD Halling-Brown","year":"2021","unstructured":"Halling-Brown, M. D. et al. OPTIMAM mammography image database: a large-scale resource of mammography images and clinical data. Radiol. Artif. Intell. 3, e200103 (2021).","journal-title":"Radiol. Artif. Intell."},{"key":"1127_CR16","doi-asserted-by":"crossref","unstructured":"Warren, L. M. et al. Deep learning to calculate breast density from processed mammography images. In Proc. 15th International Workshop on Breast Imaging (IWBI2020) (eds Bosmans, H., Marshall, N. & van Ongeval, C.) (SPIE, 2020).","DOI":"10.1117\/12.2561278"},{"key":"1127_CR17","unstructured":"Ministry of Housing, Communities and Local Government (2018 to 2021). English indices of deprivation 2019. GOV.UK https:\/\/www.gov.uk\/government\/statistics\/english-indices-of-deprivation-2019 (2019)."},{"key":"1127_CR18","unstructured":"Quinonero-Candela, J., Sugiyama, M., Schwaighofer, A. & Lawrence, N. D. (eds) Dataset Shift in Machine Learning (MIT Press, 2022)."},{"key":"1127_CR19","doi-asserted-by":"publisher","first-page":"1675","DOI":"10.1056\/NEJMoa0803545","volume":"359","author":"FJ Gilbert","year":"2008","unstructured":"Gilbert, F. J. et al. Single reading with computer-aided detection for screening mammography. N. Engl. J. Med. 359, 1675\u20131684 (2008).","journal-title":"N. Engl. J. Med."},{"key":"1127_CR20","doi-asserted-by":"publisher","first-page":"463","DOI":"10.1016\/j.acra.2011.12.016","volume":"19","author":"BD Gallas","year":"2012","unstructured":"Gallas, B. D. et al. Evaluating imaging and computer-aided detection and diagnosis devices at the FDA. Acad. Radiol. 19, 463\u2013477 (2012).","journal-title":"Acad. Radiol."},{"key":"1127_CR21","doi-asserted-by":"publisher","first-page":"8682","DOI":"10.1007\/s00330-021-07992-w","volume":"31","author":"SL van Winkel","year":"2021","unstructured":"van Winkel, S. L. et al. Impact of artificial intelligence support on accuracy and reading time in breast tomosynthesis image interpretation: a multi-reader multi-case study. Eur. Radiol. 31, 8682\u20138691 (2021).","journal-title":"Eur. Radiol."},{"key":"1127_CR22","doi-asserted-by":"publisher","DOI":"10.1186\/s12885-023-10890-7","volume":"23","author":"N Sharma","year":"2023","unstructured":"Sharma, N. et al. Multi-vendor evaluation of artificial intelligence as an independent reader for double reading in breast cancer screening on 275,900 mammograms. BMC Cancer 23, 460 (2023).","journal-title":"BMC Cancer"},{"key":"1127_CR23","doi-asserted-by":"publisher","first-page":"2850","DOI":"10.1002\/sim.6148","volume":"33","author":"MW Fagerland","year":"2014","unstructured":"Fagerland, M. W., Lydersen, S. & Laake, P. Recommended tests and confidence intervals for paired binomial proportions. Stat. Med. 33, 2850\u20132875 (2014).","journal-title":"Stat. Med."},{"key":"1127_CR24","doi-asserted-by":"publisher","first-page":"231","DOI":"10.1002\/sim.1012","volume":"21","author":"J-P Liu","year":"2002","unstructured":"Liu, J.-P., Hsueh, H.-M., Hsieh, E. & Chen, J. J. Tests for equivalence or non-inferiority for paired binary data. Stat. Med. 21, 231\u2013245 (2002).","journal-title":"Stat. Med."},{"key":"1127_CR25","doi-asserted-by":"publisher","first-page":"1495","DOI":"10.1002\/(SICI)1097-0258(19980715)17:13<1495::AID-SIM863>3.0.CO;2-I","volume":"17","author":"NA Obuchowski","year":"1998","unstructured":"Obuchowski, N. A. On the comparison of correlated proportions for clustered data. Stat. Med. 17, 1495\u20131507 (1998).","journal-title":"Stat. Med."},{"key":"1127_CR26","doi-asserted-by":"publisher","first-page":"1729","DOI":"10.1002\/(SICI)1097-0258(19960830)15:16<1729::AID-SIM334>3.0.CO;2-M","volume":"15","author":"CW Dunnett","year":"1996","unstructured":"Dunnett, C. W. & Gent, M. An alternative to the use of two-sided tests in clinical trials. Stat. Med. 15, 1729\u20131738 (1996).","journal-title":"Stat. Med."},{"key":"1127_CR27","doi-asserted-by":"publisher","first-page":"943","DOI":"10.1080\/10543400500265694","volume":"15","author":"T Koyama","year":"2005","unstructured":"Koyama, T. & Westfall, P. H. Decision-theoretic views on simultaneous testing of superiority and noninferiority. J. Biopharm. Stat. 15, 943\u2013955 (2005).","journal-title":"J. Biopharm. Stat."},{"key":"1127_CR28","doi-asserted-by":"publisher","first-page":"444","DOI":"10.1148\/radiol.2372041362","volume":"237","author":"LAL Khoo","year":"2005","unstructured":"Khoo, L. A. L., Taylor, P. & Given-Wilson, R. M. Computer-aided detection in the United Kingdom National Breast Screening Programme: prospective study. Radiology 237, 444\u2013449 (2005).","journal-title":"Radiology"}],"container-title":["Nature Cancer"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s43018-026-01127-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s43018-026-01127-0","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s43018-026-01127-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,30]],"date-time":"2026-03-30T08:55:26Z","timestamp":1774860926000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s43018-026-01127-0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,10]]},"references-count":28,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3]]}},"alternative-id":["1127"],"URL":"https:\/\/doi.org\/10.1038\/s43018-026-01127-0","relation":{},"ISSN":["2662-1347"],"issn-type":[{"value":"2662-1347","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,10]]},"assertion":[{"value":"3 October 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"16 January 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"10 March 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"Google LLC and\/or a subsidiary thereof provided the AI system and staff time to support the project. C.J.K., M.W., L.Y., M.M., R.C., J.C., J.D., D.I.G., E.G., R. Sayres, M.S., S.T., L.X., S.X. and S.S. are employees of Google and own stock as part of the standard compensation package. R. Sidebottom and F.J.G. are paid consultants of Google. L.M.W., M.H.-B., J.V., E.L., D.W. and K.Y. are employees of the Royal Surrey NHS Foundation Trust (RSNFT). RSNFT received funding from Google to support the collection of data into the OPTIMAM database. H.A. is Chief Scientific Officer, Preemptive Health, Flagship Pioneering, Chief Medical Officer, Harbinger Health and Chief Medical Officer of Oxford Medical Products. The other authors declare no competing interests.","order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}]}}