{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,31]],"date-time":"2026-10-31T19:01:03Z","timestamp":1793473263624,"version":"4.5.0"},"reference-count":51,"publisher":"BMJ","issue":"12","license":[{"start":{"date-parts":[[2025,12,31]],"date-time":"2025-12-31T00:00:00Z","timestamp":1767139200000},"content-version":"unspecified","delay-in-days":30,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["BMJ Open"],"accepted":{"date-parts":[[2025,12,17]]},"published-print":{"date-parts":[[2025,12]]},"abstract":"<jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>With growing interest in applying artificial intelligence (AI) to population breast cancer screening, the evidence base has expanded rapidly. This systematic review aims to systematically review and summarise the published evidence on the use of AI in breast cancer screening.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Design<\/jats:title>\n                    <jats:p>We conducted a systematic review of primary studies assessing AI for screening mammography, extracting test-accuracy metrics (sensitivity, specificity, recall and cancer detection rates) and workflow outcomes.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Data sources<\/jats:title>\n                    <jats:p>We searched the Cochrane Breast Cancer Group Specialised Register, Cochrane CENTRAL, PubMed, Embase (Elsevier), Scopus, ClinicalTrials.gov and the WHO International Clinical Trials Registry Platform from January 2012 to June 2025; we also screened reference lists of included studies and relevant reviews. No language restrictions were applied.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Eligibility criteria<\/jats:title>\n                    <jats:p>Primary studies evaluating AI for screening mammography (digital mammography or digital breast tomosynthesis) in asymptomatic women, assessing AI as a standalone reader or AI-assisted radiologist workflows versus radiologists alone. Eligible designs included randomised trials, prospective paired reader studies, real-world implementation\/registry cohorts, retrospective cohorts and multireader-multicase reader studies conducted in population-based or opportunistic screening settings. Key outcomes included diagnostic accuracy metrics (eg, sensitivity, specificity, Area Under the Curve (AUC) and\/or programme metrics (cancer detection rate (CDR), recall\/abnormal interpretation rate, positive predictive value, arbitration\/workload). We excluded protocols, pilot\/feasibility studies, case reports, editorials and studies without relevant accuracy or screening outcomes.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Data extraction and synthesis<\/jats:title>\n                    <jats:p>Two independent reviewers extracted data and assessed risk of bias. Study quality was appraised with Quality Assessment of Diagnostic Accuracy Studies-2 and an AI-specific critical appraisal tool, and findings were synthesised narratively with stratification by study design and AI integration role.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>31 studies met the inclusion criteria, encompassing randomised controlled trials, prospective paired-reader studies, registry-based implementations and retrospective simulations, representing more than two million screening examinations across Europe, Asia, North America and Australia. When used as a second reader or within double-reading workflows, AI generally maintained or modestly increased sensitivity (up to +9\u2009percentage points (PP)) while preserving or improving specificity. Triage and decision-referral configurations delivered the greatest operational benefit, reducing reading volumes by 40\u201390% while maintaining non-inferior cancer detection when thresholds were conservatively calibrated. Stand-alone AI achieved AUC values comparable to radiologists and similar cancer detection in real-world, non-enriched cohorts, although interval-cancer follow-up remains incomplete in several datasets. In prospective randomised evidence, including the Mammography Screening with Artificial Intelligence trial (MASAI) trial, AI-supported screening achieved higher CDRs (6.4 versus 5.0 per 1000; p=0.0021) with stable or reduced false-positive and recall rates. Across implementation and simulation settings, integration of AI reduced radiologist workload substantially, with triage and band-pass approaches reducing the number of reads by approximately 40\u201390%. Overall certainty is limited by heterogeneity across study designs, reliance on enriched datasets for some accuracy estimates and incomplete interval-cancer follow-up in several major studies.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Contemporary AI systems show diagnostic performance that is broadly comparable to radiologists and can substantially reduce reading workload, particularly when used as a second reader or triage tool. Emerging prospective evidence supports their safe integration in these roles, although transparent reporting, standardised evaluation and long-term population studies are still required before considering AI as a stand-alone reader. AI may improve workflow efficiency and possibly cancer detection, but definitive evidence on safety, especially interval cancer outcomes, remains essential.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/bmjopen-2025-111360","type":"journal-article","created":{"date-parts":[[2025,12,31]],"date-time":"2025-12-31T18:30:20Z","timestamp":1767205820000},"page":"e111360","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":12,"title":["Performance of artificial intelligence in breast cancer screening programmes: a systematic review"],"prefix":"10.1136","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6640-0558","authenticated-orcid":false,"given":"Ghufran","family":"Jassim","sequence":"first","affiliation":[{"name":"Royal College of Surgeons in Ireland and Medical University of Bahrain, Muharraq, Busaiteen, Bahrain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Osamah","family":"Otoom","sequence":"additional","affiliation":[{"name":"NHS North West, Manchester, Greater Manchester, UK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6679-5956","authenticated-orcid":false,"given":"Bindhu","family":"Nair","sequence":"additional","affiliation":[{"name":"Library & Learning Resource Centre, Royal College of Surgeons in Ireland Medical University of Bahrain, Muharraq, Busaiteen, Bahrain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jamal","family":"Hashem","sequence":"additional","affiliation":[{"name":"Royal College of Surgeons in Ireland Medical University of Bahrain, Muharraq, Bisaiteen, Bahrain"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2025,12,31]]},"reference":[{"key":"2026010703495183000_15.12.e111360.1","doi-asserted-by":"publisher","DOI":"10.3322\/caac.21708"},{"key":"2026010703495183000_15.12.e111360.2","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2009.1990"},{"key":"2026010703495183000_15.12.e111360.3","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2533090070"},{"key":"2026010703495183000_15.12.e111360.4","doi-asserted-by":"crossref","first-page":"717","DOI":"10.1016\/j.acra.2018.06.017","article-title":"Benefits of Independent Double Reading in Digital Mammography: A Theoretical Evaluation of All Possible Pairing Methodologies","volume":"26","author":"Brennan","year":"2019","journal-title":"Acad Radiol"},{"key":"2026010703495183000_15.12.e111360.5","doi-asserted-by":"crossref","first-page":"99","DOI":"10.1007\/s10462-024-10721-6","article-title":"A review of convolutional neural networks in computer vision","volume":"57","author":"Zhao","year":"2024","journal-title":"Artif Intell Rev"},{"key":"2026010703495183000_15.12.e111360.6","doi-asserted-by":"publisher","DOI":"10.3348\/kjr.2017.18.4.570"},{"key":"2026010703495183000_15.12.e111360.7","doi-asserted-by":"crossref","DOI":"10.3390\/diagnostics13010045","article-title":"The Systematic Review of Artificial Intelligence Applications in Breast Cancer Diagnosis","volume":"13","author":"Uzun Ozsahin","year":"2022","journal-title":"Diagnostics (Basel)"},{"key":"2026010703495183000_15.12.e111360.8","doi-asserted-by":"crossref","first-page":"134","DOI":"10.1002\/jmrs.385","article-title":"Artificial intelligence and convolution neural networks assessing mammographic images: a narrative literature review","volume":"67","author":"Wong","year":"2020","journal-title":"J Med Radiat Sci"},{"key":"2026010703495183000_15.12.e111360.9","doi-asserted-by":"crossref","first-page":"e106","DOI":"10.1016\/S2589-7500(20)30019-4","article-title":"Evaluating AI in breast cancer screening: a complex task","volume":"2","author":"Dustler","year":"2020","journal-title":"Lancet Digit Health"},{"key":"2026010703495183000_15.12.e111360.10","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-020-0776-9"},{"key":"2026010703495183000_15.12.e111360.11","doi-asserted-by":"crossref","first-page":"1306","DOI":"10.1001\/jamainternmed.2022.4969","article-title":"Artificial Intelligence in Breast Cancer Screening: Evaluation of FDA Device Regulation and Future Recommendations","volume":"182","author":"Potnis","year":"2022","journal-title":"JAMA Intern Med"},{"key":"2026010703495183000_15.12.e111360.12","doi-asserted-by":"crossref","first-page":"52","DOI":"10.1016\/j.imed.2023.08.001","article-title":"Medical artificial intelligence and the black box problem: a view based on the ethical principle of \u201cdo no harm\u201d","volume":"4","author":"Xu","year":"2024","journal-title":"Intelligent Medicine"},{"key":"2026010703495183000_15.12.e111360.13","doi-asserted-by":"crossref","DOI":"10.1136\/bmj.n1872","article-title":"Use of artificial intelligence for image analysis in breast cancer screening programmes: systematic review of test accuracy","volume":"374","author":"Freeman","year":"2021","journal-title":"BMJ"},{"key":"2026010703495183000_15.12.e111360.14","doi-asserted-by":"crossref","first-page":"36","DOI":"10.1016\/j.clinimag.2022.05.005","article-title":"A review of artificial intelligence in mammography","volume":"88","author":"Jairam","year":"2022","journal-title":"Clin Imaging"},{"key":"2026010703495183000_15.12.e111360.15","doi-asserted-by":"publisher","DOI":"10.1016\/S1470-2045(23)00298-X"},{"key":"2026010703495183000_15.12.e111360.16","doi-asserted-by":"crossref","first-page":"e175","DOI":"10.1016\/S2589-7500(24)00267-X","article-title":"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","volume":"7","author":"Hernstr\u00f6m","year":"2025","journal-title":"Lancet Digit Health"},{"key":"2026010703495183000_15.12.e111360.17","doi-asserted-by":"publisher","DOI":"10.1016\/S2589-7500(23)00153-X"},{"key":"2026010703495183000_15.12.e111360.18","doi-asserted-by":"crossref","DOI":"10.1038\/s41467-025-57469-3","article-title":"Artificial intelligence for breast cancer screening in mammography (AI-STREAM): preliminary analysis of a prospective multicenter cohort study","volume":"16","author":"Chang","year":"2025","journal-title":"Nat Commun"},{"key":"2026010703495183000_15.12.e111360.19","doi-asserted-by":"crossref","first-page":"917","DOI":"10.1038\/s41591-024-03408-6","article-title":"Nationwide real-world implementation of AI for cancer detection in population-based mammography screening","volume":"31","author":"Eisemann","year":"2025","journal-title":"Nat Med"},{"key":"2026010703495183000_15.12.e111360.20","doi-asserted-by":"crossref","first-page":"3958","DOI":"10.1007\/s00330-023-10426-4","article-title":"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","volume":"34","author":"El\u00edas-Cabot","year":"2024","journal-title":"Eur Radiol"},{"key":"2026010703495183000_15.12.e111360.21","doi-asserted-by":"crossref","DOI":"10.1148\/radiol.232479","article-title":"Early Indicators of the Impact of Using AI in Mammography Screening for Breast Cancer","volume":"311","author":"Lauritzen","year":"2024","journal-title":"Radiology"},{"key":"2026010703495183000_15.12.e111360.22","doi-asserted-by":"crossref","DOI":"10.1001\/jamanetworkopen.2023.35377","article-title":"APPRAISE-AI Tool for Quantitative Evaluation of AI Studies for Clinical Decision Support","volume":"6","author":"Kwong","year":"2023","journal-title":"JAMA Netw Open"},{"key":"2026010703495183000_15.12.e111360.23","doi-asserted-by":"crossref","unstructured":"Leeflang MM , Deeks JJ , eds. Cochrane handbook for systematic reviews of diagnostic test accuracy. Cochrane, 2023.","DOI":"10.1002\/14651858.ED000163"},{"key":"2026010703495183000_15.12.e111360.24","unstructured":"Covidence systematic review software. Veritas Health Innovation, Melbourne, Australia; 2025."},{"key":"2026010703495183000_15.12.e111360.25","doi-asserted-by":"publisher","DOI":"10.1059\/0003-4819-155-8-201110180-00009"},{"key":"2026010703495183000_15.12.e111360.26","doi-asserted-by":"crossref","DOI":"10.1136\/bmj.l4898","article-title":"RoB 2: a revised tool for assessing risk of bias in randomised trials","volume":"366","author":"Sterne","year":"2019","journal-title":"BMJ"},{"key":"2026010703495183000_15.12.e111360.27","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1038\/s41586-019-1799-6","article-title":"International evaluation of an AI system for breast cancer screening","volume":"577","author":"McKinney","year":"2020","journal-title":"Nature New Biol"},{"key":"2026010703495183000_15.12.e111360.28","doi-asserted-by":"crossref","DOI":"10.1148\/ryai.230033","article-title":"A Semiautonomous Deep Learning System to Reduce False Positives in Screening Mammography","volume":"6","author":"Pedemonte","year":"2024","journal-title":"Radiol Artif Intell"},{"key":"2026010703495183000_15.12.e111360.29","doi-asserted-by":"crossref","first-page":"e138","DOI":"10.1016\/S2589-7500(20)30003-0","article-title":"Changes in cancer detection and false-positive recall in mammography using artificial intelligence: a retrospective, multireader study","volume":"2","author":"Kim","year":"2020","journal-title":"Lancet Digit Health"},{"key":"2026010703495183000_15.12.e111360.30","doi-asserted-by":"publisher","DOI":"10.1007\/s12282-022-01375-9"},{"key":"2026010703495183000_15.12.e111360.31","doi-asserted-by":"crossref","first-page":"1241","DOI":"10.3348\/kjr.2022.0263","article-title":"Use of Artificial Intelligence for Reducing Unnecessary Recalls at Screening Mammography: A Simulation Study","volume":"23","author":"Kim","year":"2022","journal-title":"Korean J Radiol"},{"key":"2026010703495183000_15.12.e111360.32","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2018181371"},{"key":"2026010703495183000_15.12.e111360.33","article-title":"Implications for downstream workload based on calibrating an artificial intelligence detection algorithm by standalone-reader or combined-reader sensitivity matching","volume":"10","author":"Dembrower","year":"2023","journal-title":"J Med Imaging (Bellingham)"},{"key":"2026010703495183000_15.12.e111360.34","doi-asserted-by":"crossref","DOI":"10.1148\/ryai.230529","article-title":"AI-integrated Screening to Replace Double Reading of Mammograms: A Population-wide Accuracy and Feasibility Study","volume":"6","author":"Elhakim","year":"2024","journal-title":"Radiol Artif Intell"},{"key":"2026010703495183000_15.12.e111360.35","doi-asserted-by":"crossref","DOI":"10.1038\/s41467-024-51725-8","article-title":"Comparison of AI-integrated pathways with human-AI interaction in population mammographic screening for breast cancer","volume":"15","author":"Frazer","year":"2024","journal-title":"Nat Commun"},{"key":"2026010703495183000_15.12.e111360.36","doi-asserted-by":"crossref","first-page":"45","DOI":"10.1093\/jbi\/wbae056","article-title":"Potential Impact of an Artificial Intelligence\u2013based Mammography Triage Algorithm on Performance and Workload in a Population-based Screening Sample","volume":"7","author":"Watanabe","year":"2025","journal-title":"Journal of Breast Imaging"},{"key":"2026010703495183000_15.12.e111360.37","doi-asserted-by":"crossref","DOI":"10.1186\/s13058-024-01821-w","article-title":"Screening mammography performance according to breast density: a comparison between radiologists versus standalone intelligence detection","volume":"26","author":"Kwon","year":"2024","journal-title":"Breast Cancer Res"},{"key":"2026010703495183000_15.12.e111360.38","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2019182622"},{"key":"2026010703495183000_15.12.e111360.39","doi-asserted-by":"crossref","first-page":"e507","DOI":"10.1016\/S2589-7500(22)00070-X","article-title":"Combining the strengths of radiologists and AI for breast cancer screening: a retrospective analysis","volume":"4","author":"Leibig","year":"2022","journal-title":"Lancet Digit Health"},{"key":"2026010703495183000_15.12.e111360.40","doi-asserted-by":"crossref","first-page":"3935","DOI":"10.1007\/s00330-023-10423-7","article-title":"Population-wide evaluation of artificial intelligence and radiologist assessment of screening mammograms","volume":"34","author":"K\u00fchl","year":"2024","journal-title":"Eur Radiol"},{"key":"2026010703495183000_15.12.e111360.41","doi-asserted-by":"crossref","DOI":"10.1148\/ryai.230375","article-title":"Performance of an Artificial Intelligence System for Breast Cancer Detection on Screening Mammograms from BreastScreen Norway","volume":"6","author":"Larsen","year":"2024","journal-title":"Radiol Artif Intell"},{"key":"2026010703495183000_15.12.e111360.42","doi-asserted-by":"crossref","DOI":"10.1016\/j.ebiom.2023.104498","article-title":"Artificial intelligence (AI) for breast cancer screening: BreastScreen population-based cohort study of cancer detection","volume":"90","author":"Marinovich","year":"2023","journal-title":"EBioMedicine"},{"key":"2026010703495183000_15.12.e111360.43","doi-asserted-by":"crossref","first-page":"535","DOI":"10.1148\/radiol.211590","article-title":"Stand-Alone Use of Artificial Intelligence for Digital Mammography and Digital Breast Tomosynthesis Screening: A Retrospective Evaluation","volume":"302","author":"Romero-Mart\u00edn","year":"2022","journal-title":"Radiology"},{"key":"2026010703495183000_15.12.e111360.44","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2021203555"},{"key":"2026010703495183000_15.12.e111360.45","doi-asserted-by":"crossref","first-page":"e803","DOI":"10.1016\/S2589-7500(24)00173-0","article-title":"Strategies for integrating artificial intelligence into mammography screening programmes: a retrospective simulation analysis","volume":"6","author":"Fisches","year":"2024","journal-title":"Lancet Digit Health"},{"key":"2026010703495183000_15.12.e111360.46","doi-asserted-by":"crossref","first-page":"1184","DOI":"10.1109\/TMI.2019.2945514","article-title":"Deep Neural Networks Improve Radiologists\u2019 Performance in Breast Cancer Screening","volume":"39","author":"Wu","year":"2020","journal-title":"IEEE Trans Med Imaging"},{"key":"2026010703495183000_15.12.e111360.47","article-title":"AI-enhanced Mammography With Digital Breast Tomosynthesis for Breast Cancer Detection: Clinical Value and Comparison With Human Performance","volume":"6","author":"Resch","year":"2024","journal-title":"Radiology: Imaging Cancer"},{"key":"2026010703495183000_15.12.e111360.48","doi-asserted-by":"crossref","first-page":"100509","DOI":"10.1016\/j.ejro.2023.100509","article-title":"Artificial intelligence-based computer-assisted detection\/diagnosis (AI-CAD) for screening mammography: Outcomes of AI-CAD in the mammographic interpretation workflow","volume":"11","author":"Yoon","year":"2023","journal-title":"Eur J Radiol Open"},{"key":"2026010703495183000_15.12.e111360.49","doi-asserted-by":"crossref","first-page":"7186","DOI":"10.1007\/s00330-023-09692-z","article-title":"Effect of artificial intelligence\u2013based computer-aided diagnosis on the screening outcomes of digital mammography: a matched cohort study","volume":"33","author":"Kim","year":"2023","journal-title":"Eur Radiol"},{"key":"2026010703495183000_15.12.e111360.50","doi-asserted-by":"crossref","first-page":"258","DOI":"10.1093\/jbi\/wbad015","article-title":"Use of Artificial Intelligence for Digital Breast Tomosynthesis Screening: A Preliminary Real-world Experience","volume":"5","author":"Letter","year":"2023","journal-title":"Journal of Breast Imaging"},{"key":"2026010703495183000_15.12.e111360.51","doi-asserted-by":"publisher","DOI":"10.1007\/s00330-019-06186-9"}],"container-title":["BMJ Open"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjopen-2025-111360","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,7]],"date-time":"2026-01-07T11:54:19Z","timestamp":1767786859000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmjopen.bmj.com\/lookup\/doi\/10.1136\/bmjopen-2025-111360"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12]]},"references-count":51,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2025,12,31]]},"published-print":{"date-parts":[[2025,12]]}},"alternative-id":["10.1136\/bmjopen-2025-111360"],"URL":"https:\/\/doi.org\/10.1136\/bmjopen-2025-111360","relation":{},"ISSN":["2044-6055","2044-6055"],"issn-type":[{"value":"2044-6055","type":"print"},{"value":"2044-6055","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12]]}}}