[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-346778-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-346778-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","impact-of-using-artificial-intelligence-as-a-second-reader-in-breast-screening-including-arbitration-research-study","Impact of using artificial intelligence as a second reader in breast screening including arbitration - research study","","Incorporating artificial intelligence (AI) into a double-read breast-screening workflow with arbitration remains uncertain. This retrospective study analyzed 50,000 representative women from two NHS breast-screening centers with long-term follow-up to assess whether AI enables earlier cancer detection. For cases requiring arbitration (8,732), 22 readers applied normal arbitration workflow in a reader study. After arbitration, replacing the second reader with AI was noninferior to two human readers for sensitivity and specificity (P \u003C 0.001) and reduced workload. Arbitration improved AI specificity by overruling incorrect recalls, while occasionally overruling AI decisions for interval and next-round cancers. Further AI refinement and better explainability may support earlier cancer detection.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/impact-of-using-artificial-intelligence-as-a-second-reader-in-breast-screening-including-arbitration-research-study/346778/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/impact-of-using-artificial-intelligence-as-a-second-reader-in-breast-screening-including-arbitration-research-study/346778.png","ImageObject",300,407,{"name":42,"@type":43},"Ivy","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-11-02","2026-09-22",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",10,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"How was the AI used in the breast-screening workflow in this study?","Question",{"text":63,"@type":64},"AI was used as a second reader within a double-read workflow that also included an arbitration process for final recall decisions.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What evidence shows the AI was comparable to human readers after arbitration?",{"text":68,"@type":64},"After arbitration, replacing the second reader with AI was noninferior to two human readers for sensitivity and specificity, with P \u003C 0.001.",{"name":70,"@type":61,"acceptedAnswer":71},"What impact did arbitration have on AI recall decisions?",{"text":72,"@type":64},"Arbitration improved AI specificity by overruling some incorrect AI recalls, but it also overruled AI recall decisions for certain interval and next-round cancers.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},346778,1790924315,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":55,"slug":125},"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":139,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":144,"read_time":145,"preview_page_count":33,"preview_urls":146},549758252649,"https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819","nature cancer  \nArticle [https://doi.org/10.1038/s43018-026-01128-z](https://doi.org/10.1038/s43018-026-01128-z)  \nImpact of using artificial intelligence asa second reader in breast screening including arbitration  \nReceived: 20 December 2024  \nAccepted: 20 January 2026  \n\n| Published online: 10 March 2026 |\n| --- |\n|  Check for updates |\n\nLucy M. Warren  1 , Jenny Venton1, Kenneth C. Young  1,2,  \nMark Halling-Brown  1,2, Christopher J. Kelly  3, Marc Wilson3, MegumiMorigami3, Lisanne Khoo4, Deborah Cunningham  5,  \nRichard Sidebottom  3,6, Mamatha Reddy4, Hema Purushothaman5, Delara Khodabakhshi4, Lesley Honeyfield5, Amandeep Hujan4, TsvetinaStoycheva  5, Andy Joiner1, Reena Chopra  3, Aminata Sy7, Dominic Ward1, Lin Yang3, Rory Sayres  3, Daniel Golden  3,  \nNamrata Malhotra7, Rachita Mallya7, Lihong Xi3, Della Ogunleye8, Charlotte Purdy8, Alistair Mackenzie1, Susan Thomas9, Shravya Shetty  3  \n,  \nFiona J. Gilbert10, Ara Darzi7 & Hutan Ashrafian  7   \nThe impact of incorporating artificial intelligence (AI) into a double-read breast-screening workflow, including arbitration, is unclear. This retrospective study included 50,000 representative women from two NHS breast-screening centers. All the women had long-term follow-up, allowing us to determine whether use of AI leads to earlier cancer detection. Cases requiring arbitration (8,732 cases) were read by 22 readers ina reader study, following their normal arbitration workflow. Overall, after arbitration, replacing the second reader with AI was noninferior (5% margin) to two human readers in terms of sensitivity and specificity (P \u003C 0.001) while offering a workload benefit. Arbitration improved the specificity of the AI arm by overruling cases incorrectly recalled by the AI tool; however, it also overruled the AI tool recall decision for some interval and next-round cancers. Further development of the AI tool alongside improvement in its explainability could lead to the earlier detection of cancers.  \nBreast screening aims to find cancers early, where treatment is more successful. TheNHS Breast Screening Programme (NHSBSP) invites women aged between 50 years and 70 years for mammographic screening every 3 years, or women aged >70 years can self-refer. Each mammogram is read by two mammography readers (radiologists, breast clinicians, associate specialists, consultant radiographers or enhanced-level practitioners), who decide whether to recall the woman for additional follow-up. An arbitration process takes place  \nto come to a final decision about whether to recall a woman, either if thereis disagreement between the two mammography readers or for all recalled women, depending on local screening center processes. A radiologist workforce crisis in the UK threatens the long-term sustainability of the NHSBSP. The UK has a 30% shortfall of clinical radiologistsand this is forecast to riseto 40% by 20281. Artificial intelligence (AI)-assisted screening could play an important role in future proofing of the UK’s NHSBSP.  \n1Royal Surrey NHS Foundation Trust, Guildford, UK. 2University of Surrey, Guildford, UK. 3Google Research, Mountain View, CA, USA. 4St George’s University Hospitals NHS Foundation Trust, London, UK. 5Imperial College Healthcare NHS Trust, London, UK. 6The Royal Marsden NHS Foundation Trust, London, UK. 7Imperial College London, London, UK. 8AIMS Public Engagement Group, London, UK. 9Google for Health, Mountain View, CA, USA.  \n10University of Cambridge, Cambridge, UK. e-mail: [lucy.warren@nhs.net](lucy.warren@nhs.net); [h.ashrafian@imperial.ac.uk](h.ashrafian@imperial.ac.uk)  \na  \nHistorical screens from 50, 000 women across two screening services  \nStandard care arm  \nAI-enabled care arm  \nHistorical reader 1  \nHistorical reader 1  \nb  \nDetailed subgroup analysis across clinical, social and demographic factors  \nLarge study size allowed for subgroup analyses at new levels of detail  \nHistorical reader 2 (unblinded)  \nAI reader  \nLocal site  \narbitration rules  ","cbCaib9elHj7Pd2A","https://ap.wps.com/l/cbCaib9elHj7Pd2A","pdf",3204005,18,"English","# Study overview\n## Arbitration workflow and reader study design\n# Screening outcomes\n## Sensitivity, specificity, and workload impact\n## Interval and next-round cancer effects\n# NHS breast screening context","[{\"question\":\"How was the AI used in the breast-screening workflow in this study?\",\"answer\":\"AI was used as a second reader within a double-read workflow that also included an arbitration process for final recall decisions.\"},{\"question\":\"What evidence shows the AI was comparable to human readers after arbitration?\",\"answer\":\"After arbitration, replacing the second reader with AI was noninferior to two human readers for sensitivity and specificity, with P \\u003c 0.001.\"},{\"question\":\"What impact did arbitration have on AI recall decisions?\",\"answer\":\"Arbitration improved AI specificity by overruling some incorrect AI recalls, but it also overruled AI recall decisions for certain interval and next-round cancers.\"}]","Impact of using artificial intelligence as a second reader in breast screening including arbitration - research study | PDF",1790062710,45,[147,151,155,159],{"page_no":22,"url":148,"width":149,"height":150},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/cc4970031c6ecd81e54acb8796010686.webp",983,1305,{"page_no":26,"url":152,"width":153,"height":154},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/9233b3c5b9ac689989732c78467b10c8.webp",982,1304,{"page_no":30,"url":156,"width":157,"height":158},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/ca744b8a31b6563337e867367811cd1d.webp",981,1303,{"page_no":33,"url":160,"width":161,"height":162},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/9b864bafe8c3bc9ab9660e7285b8a1bb.webp",980,1301]