[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-351337-105":59,"doc-detail-351337-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","prospective-evaluation-of-artificial-intelligence-integration-into-breast-cancer-screening-in-multiple-workflow-settings-the-gemini-study","Prospective evaluation of artificial intelligence integration into breast cancer screening in multiple workflow settings: the GEMINI study","","Prospective service evaluation of the GEMINI study assessed how integrating an artificial intelligence tool into breast cancer screening workflows affects clinical outcomes and operational workload. In one UK region, 10,889 women received routine care while the study evaluated 17 AI implementation configurations using live integration and workflow simulations. When AI recommended recall beyond routine double reading, additional human review detected 11 more cancers. The primary workflow increased cancer detection by 10.4%, maintained recall rate, and reduced workload by up to 31%, with other configurations improving cancer metrics and achieving up to 36% savings.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/prospective-evaluation-of-artificial-intelligence-integration-into-breast-cancer-screening-in-multiple-workflow-settings-the-gemini-study/351337/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/prospective-evaluation-of-artificial-intelligence-integration-into-breast-cancer-screening-in-multiple-workflow-settings-the-gemini-study/351337.png","ImageObject",300,407,{"name":92,"@type":93},"\tCallum ","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-17","2026-09-22",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",14,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"How was AI integrated into the breast cancer screening workflows in the GEMINI study?","Question",{"text":113,"@type":114},"The study used both live AI integration within a routine service and workflow simulations. It assessed AI as an additional read for suspicious cases and as a triage approach to reduce workload.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"What happened when AI recommended recall beyond routine double reading?",{"text":118,"@type":114},"Cases where AI recommended recall but routine double reading did not received additional human review, leading to detection of 11 additional cancers.",{"name":120,"@type":111,"acceptedAnswer":121},"What were the reported effects on cancer detection, recall rate, and workload?",{"text":122,"@type":114},"The primary AI workflow improved cancer detection by 10.4% (about 1 per 1,000), maintained the recall rate with only a 0.8% reduction, and reduced workload by up to 31%. Other workflow variations showed improvements across multiple measured metrics with up to 36% workload savings.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},351337,1791362761,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":145,"read_time":146,"preview_page_count":19,"preview_urls":147},137451211410,"https://ap-avatar.wpscdn.com/avatar/2000bb0a9246f588df?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786362646172706240","nature cancer  \nArticle [https://doi.org/10.1038/s43018-026-01126-1](https://doi.org/10.1038/s43018-026-01126-1)  \nProspective evaluation of artificial intelligence integration into breast cancer screening in multiple workflow settings: the GEMINI study  \nReceived: 1 October 2024  \nAccepted: 16 January 2026  \n\n| Published online: 10 March 2026 |\n| --- |\n|  Check for updates |\n\nClarisse Florence de Vries  1 , Gerald Lip  2, Roger Todd Staff2, Jaroslaw Artur Dymiter3, Benjamin Tse2, Annie Ng  4, Georgia Fox5, Cary Oberije  5 & Lesley Ann Anderson  6,7  \nArtificial intelligence (AI) tools can improve breast screening performance but different screening sites have varying needs. Here the GEMINI prospective evaluation of 10,889 women, within one UK region, used both live AI integration and simulations to model 17 different ways AI could be used in breast screening. All women received routine care. One AI tool was assessed. When the AI tool recommended recall but routine double reading did not, cases underwent additional human review, detecting 11 additional cancers. The primary AI workflow could improve cancer detection by 10.4%(1 per 1,000), maintain the recall rate (0.8% reduction) and reduce workload by up to 31%. Other workflow variations significantly improved all measured metrics (superiority in cancer detection rate, recall rate, positive predictive value (PPV), sensitivity and specificity) with up to 36% workload savings. Different AI integrationsin breast screening could offer various clinical and operational gains, allowing for adaptation to local healthcare needs.  \nArtificial intelligence (AI) tools improve breast screening by reducing workload and maintaining or improving cancer detection rates (CDRs)1–4. AIisnow being used in some clinics in the United States5, Hungary4 and within the Danish Capital Region breast cancer screening program3.  \nMost studies evaluating AI performance have used historical data6,7. These retrospective studies have limited ability to assess how AI may influence clinical practice, and specifically, improve cancer detection. Recent prospective studies have demonstrated that AI use in breast screening can be safe and effective, increasing cancer detection and reducing workload1,2,8. The Swedish ScreenTrustCAD study reported that double reading by one radiologist plus AI resulted  \nin 4% more screen-detected cancers and a 4% lower recall rate than routine double human reading1,2,8. The Swedish MASAI randomized controlled trial reported in its clinical safety analysis (n = 80,033) that AI-supported screening (single or double reading based on AI risk score) detected 1 per 1,000 more cancers and reduced workload by 44.3% compared to routine screening8. Similarly in Denmark, AI-assisted reading increased CDR by 1 per 1,000, while reducing therecall rate by 20.5% and reading workload by 33.5%3.  \nDespite these promising results, the demonstrated improvements depend on the AI tooland how it is integrated into the screening pathway. Randomized controlled trial methods limit assessment to  \n1Glasgow Lab for Data Science & AI, Public Health, School of Health and Wellbeing, University of Glasgow, Glasgow, UK. 2National Health Service Grampian, Aberdeen, UK. 3Grampian Data Safe Haven (DaSH), University of Aberdeen, Aberdeen, UK. 4DeepHealth Inc., Somerville, MA, USA. 5Kheiron Medical Technologies Ltd., London, UK. 6Biostatistics and Health Data Science, Institute of Applied Health Sciences, School of Medicine, Medical Sciences & Nutrition, University of Aberdeen, Aberdeen, UK. 7Interdisciplinary Institute, University of Aberdeen, Aberdeen, UK.  \n[e-mail:](e-mail: clarisse.devries@glasgow.ac.uk)[ clarisse.devries@glasgow.ac.uk](e-mail: clarisse.devries@glasgow.ac.uk)  \nFig. 1 | The diagram shows the flow of screening attendees through to study inclusion. Blue boxes indicate retained participants; white boxes indicate exclusions.  \none or two implementation strategies, while AI can be used in various config","cbCainRgTL71dwRi","https://ap.wps.com/l/cbCainRgTL71dwRi","pdf",1370446,13,"English","# Prospective evaluation overview\n## Study design and AI integration strategy\n## AI tool assessment approaches\n## Outcomes: cancer detection, recall rate, workload\n## Cohort characteristics and demographics","[{\"question\":\"How was AI integrated into the breast cancer screening workflows in the GEMINI study?\",\"answer\":\"The study used both live AI integration within a routine service and workflow simulations. It assessed AI as an additional read for suspicious cases and as a triage approach to reduce workload.\"},{\"question\":\"What happened when AI recommended recall beyond routine double reading?\",\"answer\":\"Cases where AI recommended recall but routine double reading did not received additional human review, leading to detection of 11 additional cancers.\"},{\"question\":\"What were the reported effects on cancer detection, recall rate, and workload?\",\"answer\":\"The primary AI workflow improved cancer detection by 10.4% (about 1 per 1,000), maintained the recall rate with only a 0.8% reduction, and reduced workload by up to 31%. Other workflow variations showed improvements across multiple measured metrics with up to 36% workload savings.\"}]","Prospective evaluation of artificial intelligence integration into breast cancer screening in multiple workflow settings: the GEMINI study | PDF",1790093597,33,[148,152,156,160],{"page_no":8,"url":149,"width":150,"height":151},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/26d4f2ca4a5f1feb09b9374c26357c68.webp",983,1305,{"page_no":14,"url":153,"width":154,"height":155},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/d856540131ba96fbf81df7863af21bc5.webp",982,1304,{"page_no":81,"url":157,"width":158,"height":159},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/b018229482ea1da80fac3e0b572d156e.webp",981,1303,{"page_no":19,"url":161,"width":162,"height":163},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/c72ddc68d87fc41e4584f3fdaee7f091.webp",980,1301]