{"version":"6.9","hitCount":5,"request":{"queryString":"PMCID:PMC13037975 OR PMCID:PMC13405297 OR PMCID:PMC13377025 OR PMCID:PMC13524940 OR PMCID:PMC13492657","resultType":"core","cursorMark":"*","pageSize":25,"sort":"","synonym":false},"resultList":{"result":[{"id":"42623551","source":"MED","pmid":"42623551","pmcid":"PMC13492657","fullTextIdList":{"fullTextId":["PMC13492657"]},"doi":"10.1093/esj/aakag098","title":"Implementation of an AI-supported decision-making tool in a high-volume stroke system with routine perfusion imaging.","authorString":"Karlsson A, Cadeborn E, Woock M, Allardt A, Jood K, Björkman-Burtscher I, Rentzos A.","authorList":{"author":[{"fullName":"Karlsson A","lastName":"Karlsson","firstName":"Adrian","initials":"A","authorId":{"type":"ORCID","value":"0000-0001-5428-9995"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Radiology, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden."},{"affiliation":"Section of Diagnostic and Interventional Neuroradiology, Department of Radiology, Sahlgrenska University Hospital, Region Västra Götaland, Gothenburg, Sweden."}]}},{"fullName":"Cadeborn E","lastName":"Cadeborn","firstName":"Erik","initials":"E","authorId":{"type":"ORCID","value":"0009-0002-1225-7359"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Section of Diagnostic and Interventional Neuroradiology, Department of Radiology, Sahlgrenska University Hospital, Region Västra Götaland, Gothenburg, Sweden."}]}},{"fullName":"Woock M","lastName":"Woock","firstName":"Malin","initials":"M","authorId":{"type":"ORCID","value":"0000-0002-9291-3151"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Clinical Neuroscience, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden."},{"affiliation":"Department of Neurology, Sahlgrenska University Hospital, Region Västra Götaland, Gothenburg, Sweden."}]}},{"fullName":"Allardt A","lastName":"Allardt","firstName":"Arne","initials":"A","authorId":{"type":"ORCID","value":"0009-0000-9142-0737"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Clinical Neuroscience, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden."},{"affiliation":"Department of Neurology, Sahlgrenska University Hospital, Region Västra Götaland, Gothenburg, Sweden."}]}},{"fullName":"Jood K","lastName":"Jood","firstName":"Katarina","initials":"K","authorId":{"type":"ORCID","value":"0000-0001-8746-1771"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Clinical Neuroscience, Institute of Neuroscience and Physiology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden."},{"affiliation":"Department of Neurology, Sahlgrenska University Hospital, Region Västra Götaland, Gothenburg, Sweden."}]}},{"fullName":"Björkman-Burtscher I","lastName":"Björkman-Burtscher","firstName":"Isabella","initials":"I","authorId":{"type":"ORCID","value":"0000-0002-9023-3363"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Radiology, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden."},{"affiliation":"Section of Diagnostic and Interventional Neuroradiology, Department of Radiology, Sahlgrenska University Hospital, Region Västra Götaland, Gothenburg, Sweden."}]}},{"fullName":"Rentzos A","lastName":"Rentzos","firstName":"Alexandros","initials":"A","authorId":{"type":"ORCID","value":"0000-0001-6610-7802"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Radiology, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden."},{"affiliation":"Section of Diagnostic and Interventional Neuroradiology, Department of Radiology, Sahlgrenska University Hospital, Region Västra Götaland, Gothenburg, Sweden."}]}}]},"authorIdList":{"authorId":[{"type":"ORCID","value":"0000-0001-5428-9995"},{"type":"ORCID","value":"0000-0001-6610-7802"},{"type":"ORCID","value":"0000-0001-8746-1771"},{"type":"ORCID","value":"0000-0002-9023-3363"},{"type":"ORCID","value":"0000-0002-9291-3151"},{"type":"ORCID","value":"0009-0000-9142-0737"},{"type":"ORCID","value":"0009-0002-1225-7359"}]},"journalInfo":{"issue":"8","volume":"11","journalIssueId":4240880,"dateOfPublication":"2026 Aug","monthOfPublication":8,"yearOfPublication":2026,"journal":{"medlineAbbreviation":"Eur Stroke J","title":"European stroke journal","essn":"2396-9881","issn":"2396-9873","nlmid":"101688446","isoabbreviation":"Eur Stroke J"},"printPublicationDate":"2026-08-01"},"pubYear":"2026","pageInfo":"aakag098","abstractText":"<h4>Introduction</h4>Endovascular thrombectomy (EVT) is the standard of care for LVO stroke; however, rapid workflow is critical for a beneficial outcome. Artificial intelligence (AI)-supported imaging tools, such as Brainomix 360 Stroke, may improve the efficiency of acute stroke imaging and reduce treatment delays. We evaluated the effects of regional implementation of Brainomix 360 Stroke in a high-volume system with routine use of perfusion imaging.<h4>Patients and methods</h4>We conducted a retrospective register-based cohort study of consecutive patients treated with EVT at Sahlgrenska University Hospital from June 2021 to May 2024. Patients outside the Region Västra Götaland prehospital area or aged under 18 years were excluded. The study period was divided into pre-implementation, learning and established periods following Brainomix 360 Stroke introduction. Primary organisational outcomes were time from non-contrast brain CT (NCCT) to CT perfusion map availability (CT-to-perfusion-map-availability) and from NCCT to groin puncture (CT-puncture). Clinical outcomes included early neurological improvement (≥4 points on the NIHSS or a score of 0-1 at 24 h) and favourable functional outcome (mRS scores 0-2 or return to pre-stroke mRS at 90 days).<h4>Results</h4>A total of 970 EVT patients were included. Adjusted median CT-puncture significantly decreased from 47 min pre-implementation to 36 and 35 min in the learning and established periods, respectively. Adjusted median CT-to-perfusion-map-availability significantly decreased from 7 min pre-implementation to 6 min in the established period. In subgroup analyses of primary stroke centres (PSCs), results were similar while no significant differences were found in subgroup analyses of the comprehensive stroke centre. No significant differences were observed in clinical outcomes.<h4>Discussion and conclusion</h4>Implementation of an AI-supported imaging decision-making tool was associated with significant reductions in key workflow times, largely attributable to decreased times at PSCs in the region.","affiliation":"Department of Radiology, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.","publicationStatus":"ppublish","language":"eng","pubModel":"Print","pubTypeList":{"pubType":["research-article","Journal Article"]},"meshHeadingList":{"meshHeading":[{"majorTopic_YN":"N","descriptorName":"Humans"},{"majorTopic_YN":"N","descriptorName":"Tomography, X-Ray Computed"},{"majorTopic_YN":"N","descriptorName":"Thrombectomy","meshQualifierList":{"meshQualifier":[{"majorTopic_YN":"N","abbreviation":"MT","qualifierName":"methods"}]}},{"majorTopic_YN":"N","descriptorName":"Registries"},{"majorTopic_YN":"N","descriptorName":"Retrospective Studies"},{"majorTopic_YN":"Y","descriptorName":"Artificial Intelligence"},{"majorTopic_YN":"N","descriptorName":"Aged"},{"majorTopic_YN":"N","descriptorName":"Aged, 80 and over"},{"majorTopic_YN":"N","descriptorName":"Middle Aged"},{"majorTopic_YN":"N","descriptorName":"Female"},{"majorTopic_YN":"N","descriptorName":"Male"},{"majorTopic_YN":"Y","descriptorName":"Stroke","meshQualifierList":{"meshQualifier":[{"majorTopic_YN":"N","abbreviation":"DG","qualifierName":"diagnostic imaging"},{"majorTopic_YN":"N","abbreviation":"SU","qualifierName":"surgery"}]}},{"majorTopic_YN":"Y","descriptorName":"Perfusion Imaging","meshQualifierList":{"meshQualifier":[{"majorTopic_YN":"N","abbreviation":"MT","qualifierName":"methods"}]}},{"majorTopic_YN":"N","descriptorName":"Endovascular Procedures","meshQualifierList":{"meshQualifier":[{"majorTopic_YN":"N","abbreviation":"MT","qualifierName":"methods"}]}},{"majorTopic_YN":"Y","descriptorName":"Clinical Decision-Making"}]},"keywordList":{"keyword":["Artificial intelligence","Stroke","Endovascular treatment"]},"subsetList":{"subset":[{"code":"IM","name":"Index Medicus"}]},"fullTextUrlList":{"fullTextUrl":[{"availability":"Subscription required","availabilityCode":"S","documentStyle":"doi","site":"DOI","url":"https://doi.org/10.1093/esj/aakag098"},{"availability":"Open access","availabilityCode":"OA","documentStyle":"html","site":"Europe_PMC","url":"https://europepmc.org/articles/PMC13492657"},{"availability":"Open access","availabilityCode":"OA","documentStyle":"pdf","site":"Europe_PMC","url":"https://europepmc.org/articles/PMC13492657?pdf=render"}]},"isOpenAccess":"Y","inEPMC":"Y","inPMC":"Y","citedByCount":0,"hasReferences":"Y","hasTextMinedTerms":"Y","hasDbCrossReferences":"N","hasPDF":"Y","hasBook":"N","hasSuppl":"Y","hasLabsLinks":"N","hasData":"Y","license":"cc by","authMan":"N","epmcAuthMan":"N","nihAuthMan":"N","hasTMAccessionNumbers":"N","dateOfCompletion":"2026-08-20","dateOfCreation":"2026-08-20","dateOfRevision":"2026-08-22","firstPublicationDate":"2026-08-01","firstIndexDate":"2026-08-22","fullTextReceivedDate":"2026-09-09","hasEvaluations":"N"},{"id":"42507719","source":"MED","pmid":"42507719","pmcid":"PMC13405297","fullTextIdList":{"fullTextId":["PMC13405297"]},"doi":"10.1371/journal.pdig.0001534","title":"Validation is not enough: Longitudinal evidence of post-deployment fragility in clinical AI systems.","authorString":"Kopanitsa G.","authorList":{"author":[{"fullName":"Kopanitsa G","lastName":"Kopanitsa","firstName":"Georgy","initials":"G","authorId":{"type":"ORCID","value":"0000-0002-6231-8036"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Federal State Budgetary Institution, V.A. Almazov National Medical Research Centre of the Ministry of Health of the Russian Federation, Saint-Petersburg, Russia."},{"affiliation":"ITMO University, Saint-Petersburg, Russia."}]}}]},"authorIdList":{"authorId":[{"type":"ORCID","value":"0000-0002-6231-8036"}]},"dataLinksTagsList":{"dataLinkstag":["supporting_data"]},"journalInfo":{"issue":"7","volume":"5","journalIssueId":4223981,"dateOfPublication":"2026 Jul","monthOfPublication":7,"yearOfPublication":2026,"journal":{"medlineAbbreviation":"PLOS Digit Health","title":"PLOS digital health","essn":"2767-3170","issn":"2767-3170","nlmid":"9918335064206676","isoabbreviation":"PLOS Digit Health"},"printPublicationDate":"2026-07-01"},"pubYear":"2026","pageInfo":"e0001534","abstractText":"Pre-deployment validation is commonly used to establish the safety and effectiveness of clinical artificial intelligence systems, but acceptable validation performance does not guarantee stable behavior after deployment into routine clinical workflows. We conducted a longitudinal retrospective observational study of four clinically deployed AI systems operating across distinct clinical domains and workflows within a large healthcare organization. Using routinely collected clinical data, outcome labels, and operational telemetry, we compared validation-era performance with post-deployment behavior over extended observation periods. Analyses focused on temporal patterns of discrimination, calibration, data availability, latency, and workflow-related signals, with particular attention to label-dependent and label-independent monitoring. Across all systems, validation-era performance did not persist as a stable operational property after deployment. Calibration drift emerged consistently and often preceded detectable changes in discrimination. Workflow-associated changes in data availability and timing were more strongly and consistently associated with degradation than population-level indicators. Label-independent operational signals, including input missingness and data latency, provided early indication of emerging fragility, whereas outcome-based monitoring was delayed by label latency and documentation processes. These findings suggest that post-deployment fragility can be a structural property of clinical AI systems embedded in evolving workflows. Effective governance therefore requires lifecycle-oriented monitoring strategies that combine calibration reassessment with operational telemetry throughout deployment.","affiliation":"Federal State Budgetary Institution, V.A. Almazov National Medical Research Centre of the Ministry of Health of the Russian Federation, Saint-Petersburg, Russia.","publicationStatus":"epublish","language":"eng","pubModel":"Electronic-eCollection","pubTypeList":{"pubType":["research-article","Journal Article"]},"fullTextUrlList":{"fullTextUrl":[{"availability":"Subscription required","availabilityCode":"S","documentStyle":"doi","site":"DOI","url":"https://doi.org/10.1371/journal.pdig.0001534"},{"availability":"Open access","availabilityCode":"OA","documentStyle":"html","site":"Europe_PMC","url":"https://europepmc.org/articles/PMC13405297"},{"availability":"Open access","availabilityCode":"OA","documentStyle":"pdf","site":"Europe_PMC","url":"https://europepmc.org/articles/PMC13405297?pdf=render"}]},"isOpenAccess":"Y","inEPMC":"Y","inPMC":"Y","citedByCount":0,"hasReferences":"Y","hasTextMinedTerms":"Y","hasDbCrossReferences":"N","hasPDF":"Y","hasBook":"N","hasSuppl":"Y","hasLabsLinks":"N","hasData":"Y","license":"cc by","authMan":"N","epmcAuthMan":"N","nihAuthMan":"N","hasTMAccessionNumbers":"Y","tmAccessionTypeList":{"accessionType":["doi"]},"dateOfCompletion":"2026-07-27","dateOfCreation":"2026-07-27","dateOfRevision":"2026-08-13","electronicPublicationDate":"2026-07-27","firstPublicationDate":"2026-07-27","firstIndexDate":"2026-07-30","fullTextReceivedDate":"2026-09-04","hasEvaluations":"N"},{"id":"42410010","source":"MED","pmid":"42410010","pmcid":"PMC13524940","fullTextIdList":{"fullTextId":["PMC13524940"]},"doi":"10.1038/s41598-026-61073-w","title":"Implementation of an ai-enabled multimodal emergency care system is associated with improved sudden cardiac death rescue outcomes in anyang.","authorString":"Liu X, Zhang C, Zhang H, Shen J, Liu P, Wang Y, Li R, Zhang S.","authorList":{"author":[{"fullName":"Liu X","lastName":"Liu","firstName":"Xiaopeng","initials":"X","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Emergency Medicine, The Fifth Clinical Medical College of Henan University of Chinese Medicine, Zhengzhou, 450002, Henan Province, China."}]}},{"fullName":"Zhang C","lastName":"Zhang","firstName":"Chenlong","initials":"C","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"School of Cyberspace Security, The PLA Information Engineering University, Zhengzhou, 450000, Henan Province, China."}]}},{"fullName":"Zhang H","lastName":"Zhang","firstName":"Hongjiang","initials":"H","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Emergency Medicine, Hua County People's Hospital, Anyang, 456400, Henan Province, China."}]}},{"fullName":"Shen J","lastName":"Shen","firstName":"Junyang","initials":"J","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Beijing Yunhui Medical Technology Co., Ltd, Beijing, 100000, China."}]}},{"fullName":"Liu P","lastName":"Liu","firstName":"Peng","initials":"P","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Physics, State Key Laboratory of Low-Dimensional Quantum Physics, Tsinghua University, Tsinghua University, Beijing, 100084, China."}]}},{"fullName":"Wang Y","lastName":"Wang","firstName":"Yunlong","initials":"Y","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Henan Bioengineering Technology Research Center, Zhengzhou, 450018, Henan Province, China."}]}},{"fullName":"Li R","lastName":"Li","firstName":"Rui","initials":"R","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Emergency Medicine, Zhongmu County People's Hospital, Zhengzhou, Henan Province, China. LJZ66@163.com."}]}},{"fullName":"Zhang S","lastName":"Zhang","firstName":"Sisen","initials":"S","authorId":{"type":"ORCID","value":"0000-0001-6090-2290"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Emergency Medicine, The Fifth Clinical Medical College of Henan University of Chinese Medicine, Zhengzhou, 450002, Henan Province, China. zhangsisen@hactcm.edu.cn."},{"affiliation":"Department of Emergency Medicine, Zhengzhou People's Hospital, Zhengzhou, 450003, China. zhangsisen@hactcm.edu.cn."}]}}]},"authorIdList":{"authorId":[{"type":"ORCID","value":"0000-0001-6090-2290"}]},"dataLinksTagsList":{"dataLinkstag":["supporting_data"]},"journalInfo":{"issue":"1","volume":"16","journalIssueId":4223254,"dateOfPublication":"2026 Jul","monthOfPublication":7,"yearOfPublication":2026,"journal":{"medlineAbbreviation":"Sci Rep","title":"Scientific reports","essn":"2045-2322","issn":"2045-2322","nlmid":"101563288","isoabbreviation":"Sci Rep"},"printPublicationDate":"2026-07-01"},"pubYear":"2026","pageInfo":"27092","abstractText":"Sudden Cardiac Death (SCD) remains a leading cause of mortality worldwide, with outcomes critically dependent on the effective implementation of the \"Chain of Survival\" - early recognition, early CPR, early defibrillation, and post-resuscitation care. In regional and pre-hospital settings, systemic fragmentation between emergency dispatch, ambulance services, and hospitals undermines this chain. This study presents the development, implementation, and impact evaluation of an integrated, AI-enabled multi-modal emergency care system designed to strengthen the entire Chain of Survival for SCD in a regional context.: We designed and deployed a system integrating a unified information platform, IoT-enabled devices, point-of-care testing (POCT), and AI-driven clinical decision support. The system was implemented phased across three counties in Anyang, China (population ≈ 2.1 million) from January 2022 to December 2023. We conducted a quasi-experimental before-and-after study using routinely collected emergency medical services (EMS) data. Primary outcomes were median response time (call receipt to scene arrival), pre-hospital STEMI identification rate, and return of spontaneous circulation (ROSC) for out-of-hospital cardiac arrest (OHCA) of cardiac origin. Data from 1,208 emergency cases (pre-implementation: n = 587; post-implementation: n = 621) were analyzed. Interrupted time series (ITS) analysis was performed to control for secular trends. The median emergency response time decreased from 9.8 min (IQR: 7.2-13.1) to 6.7 min (IQR: 5.1-9.0) (P < 0.001). The pre-hospital STEMI identification rate improved from 65% to 90% (p < 0.01). For OHCA of cardiac origin, the ROSC rate increased from 18% to 31% (p < 0.05), representing a 72% relative improvement. ITS analysis confirmed a significant level change for response time (β = -2.8 min, 95% CI: -3.7 to -1.9, P < 0.001) and for ROSC (β = +12% points, 95% CI: +5 to + 19, P = 0.01) immediately following implementation, with no significant pre-existing trends. The AI models demonstrated robust performance during validation (deterioration prediction AUC 0.89; STEMI detection AUC 0.92). The Anyang Model provides evidence that a systematically integrated, AI-driven platform is feasible and temporally associated with substantial improvements in regional emergency care for SCD. While causal attribution requires further validation, this systems-level approach offers a replicable framework that can be adapted to diverse resource settings.","affiliation":"Department of Emergency Medicine, The Fifth Clinical Medical College of Henan University of Chinese Medicine, Zhengzhou, 450002, Henan Province, China.","publicationStatus":"epublish","language":"eng","pubModel":"Electronic","pubTypeList":{"pubType":["research-article","Journal Article"]},"grantsList":{"grant":[{"agency":"Anyang City Major Science and Technology Special Projec","grantId":"2025A02SF008","orderIn":0},{"agency":"Natural Science Foundation of Henan Province","grantId":"232300420059","orderIn":0}]},"meshHeadingList":{"meshHeading":[{"majorTopic_YN":"N","descriptorName":"Humans"},{"majorTopic_YN":"Y","descriptorName":"Death, Sudden, Cardiac","meshQualifierList":{"meshQualifier":[{"majorTopic_YN":"N","abbreviation":"EP","qualifierName":"epidemiology"},{"majorTopic_YN":"N","abbreviation":"PC","qualifierName":"prevention & control"}]}},{"majorTopic_YN":"N","descriptorName":"Cardiopulmonary Resuscitation","meshQualifierList":{"meshQualifier":[{"majorTopic_YN":"N","abbreviation":"MT","qualifierName":"methods"}]}},{"majorTopic_YN":"Y","descriptorName":"Artificial Intelligence"},{"majorTopic_YN":"Y","descriptorName":"Emergency Medical Services","meshQualifierList":{"meshQualifier":[{"majorTopic_YN":"N","abbreviation":"MT","qualifierName":"methods"}]}},{"majorTopic_YN":"N","descriptorName":"Female"}]},"keywordList":{"keyword":["Artificial intelligence","Emergency Medicine","Clinical Decision Support System","Sudden Cardiac Death","Point-of-care Testing","Internet Of Things","Multi-modal Data","Regional Emergency System"]},"subsetList":{"subset":[{"code":"IM","name":"Index Medicus"}]},"fullTextUrlList":{"fullTextUrl":[{"availability":"Subscription required","availabilityCode":"S","documentStyle":"doi","site":"DOI","url":"https://doi.org/10.1038/s41598-026-61073-w"},{"availability":"Open access","availabilityCode":"OA","documentStyle":"html","site":"Europe_PMC","url":"https://europepmc.org/articles/PMC13524940"},{"availability":"Open access","availabilityCode":"OA","documentStyle":"pdf","site":"Europe_PMC","url":"https://europepmc.org/articles/PMC13524940?pdf=render"}]},"isOpenAccess":"Y","inEPMC":"Y","inPMC":"Y","citedByCount":0,"hasReferences":"Y","hasTextMinedTerms":"Y","hasDbCrossReferences":"N","hasPDF":"Y","hasBook":"N","hasSuppl":"Y","hasLabsLinks":"N","hasData":"Y","license":"cc by","authMan":"N","epmcAuthMan":"N","nihAuthMan":"N","hasTMAccessionNumbers":"Y","tmAccessionTypeList":{"accessionType":["doi"]},"dateOfCompletion":"2026-08-28","dateOfCreation":"2026-07-06","dateOfRevision":"2026-08-31","electronicPublicationDate":"2026-07-06","firstPublicationDate":"2026-07-06","firstIndexDate":"2026-07-08","fullTextReceivedDate":"2026-09-10","hasEvaluations":"N"},{"id":"42120428","source":"MED","pmid":"42120428","pmcid":"PMC13377025","fullTextIdList":{"fullTextId":["PMC13377025"]},"doi":"10.1038/s41467-026-72960-1","title":"Artificial intelligence for predicting hospital admissions from the emergency department: a prospective, quasi-experimental study.","authorString":"Ryu AJ, Ayanian S, Qian R, Parikh RS, Dugani SB, Fischer KM, Heaton HA, Boyum JP, Hinton BJ, Lawson DK, Burton MC.","authorList":{"author":[{"fullName":"Ryu AJ","lastName":"Ryu","firstName":"Alexander J","initials":"AJ","authorId":{"type":"ORCID","value":"0000-0002-0138-5112"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Division of Hospital Internal Medicine, Mayo Clinic, Rochester, MN, USA. Ryu.alexander@mayo.edu."}]}},{"fullName":"Ayanian S","lastName":"Ayanian","firstName":"Shant","initials":"S","authorId":{"type":"ORCID","value":"0000-0001-9319-9001"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Division of Hospital Internal Medicine, Mayo Clinic, Rochester, MN, USA."}]}},{"fullName":"Qian R","lastName":"Qian","firstName":"Ray","initials":"R","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Division of Hospital Internal Medicine, Mayo Clinic, Rochester, MN, USA."}]}},{"fullName":"Parikh RS","lastName":"Parikh","firstName":"Riddhi S","initials":"RS","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Division of Hospital Internal Medicine, Mayo Clinic, Rochester, MN, USA."}]}},{"fullName":"Dugani SB","lastName":"Dugani","firstName":"Sagar B","initials":"SB","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Division of Hospital Internal Medicine, Mayo Clinic, Rochester, MN, USA."}]}},{"fullName":"Fischer KM","lastName":"Fischer","firstName":"Karen M","initials":"KM","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Quantitative Health Sciences, Mayo Clinic, Rochester, MN, USA."}]}},{"fullName":"Heaton HA","lastName":"Heaton","firstName":"Heather A","initials":"HA","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Emergency Medicine, Mayo Clinic, Rochester, MN, USA."}]}},{"fullName":"Boyum JP","lastName":"Boyum","firstName":"Jens P","initials":"JP","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Clinical Systems, Mayo Clinic, Rochester, MN, USA."}]}},{"fullName":"Hinton BJ","lastName":"Hinton","firstName":"Benjamin J","initials":"BJ","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Center for Digital Health, Mayo Clinic, Rochester, MN, USA."}]}},{"fullName":"Lawson DK","lastName":"Lawson","firstName":"Donna K","initials":"DK","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Division of Hospital Internal Medicine, Mayo Clinic, Rochester, MN, USA."}]}},{"fullName":"Burton MC","lastName":"Burton","firstName":"M Caroline","initials":"MC","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Division of Hospital Internal Medicine, Mayo Clinic, Rochester, MN, USA."}]}}]},"authorIdList":{"authorId":[{"type":"ORCID","value":"0000-0001-9319-9001"},{"type":"ORCID","value":"0000-0002-0138-5112"}]},"dataLinksTagsList":{"dataLinkstag":["altmetrics","supporting_data"]},"journalInfo":{"issue":"1","volume":"17","journalIssueId":4187584,"dateOfPublication":"2026 May","monthOfPublication":5,"yearOfPublication":2026,"journal":{"medlineAbbreviation":"Nat Commun","title":"Nature communications","essn":"2041-1723","issn":"2041-1723","nlmid":"101528555","isoabbreviation":"Nat Commun"},"printPublicationDate":"2026-05-01"},"pubYear":"2026","pageInfo":"6337","abstractText":"The use of certain artificial intelligence (AI) tools may improve hospital operational efficiency, in particular in overcrowded emergency departments (ED). Here, we conduct a prospective, quasi experimental study evaluating an AI model predicting hospital admission risk, alternately displaying and hiding its outputs to clinicians in 2 week blocks over 11 months in 2023. Among 54,394 eligible ED visits, the AI tool does not change the number of ED patients discharged per day but reduces median ED length of stay by 12 min without increasing 72 h bounceback visits. Model performance remained stable (AUC 0.80-0.82), and hospitalist clinicians reported greater perceived usefulness than ED clinicians. These findings show that integrating a low burden AI prediction tool into ED workflows can improve operational efficiency. The study was registered on Clinicaltrials.gov (NCT05683899).","affiliation":"Division of Hospital Internal Medicine, Mayo Clinic, Rochester, MN, USA. Ryu.alexander@mayo.edu.","publicationStatus":"epublish","language":"eng","pubModel":"Electronic","pubTypeList":{"pubType":["research-article","Journal Article"]},"grantsList":{"grant":[{"agency":"NIMHD NIH HHS","acronym":"MD","grantId":"K23 MD016230","orderIn":0}]},"meshHeadingList":{"meshHeading":[{"majorTopic_YN":"N","descriptorName":"Humans"},{"majorTopic_YN":"Y","descriptorName":"Hospitalization","meshQualifierList":{"meshQualifier":[{"majorTopic_YN":"N","abbreviation":"SN","qualifierName":"statistics & numerical data"}]}},{"majorTopic_YN":"N","descriptorName":"Length of Stay","meshQualifierList":{"meshQualifier":[{"majorTopic_YN":"N","abbreviation":"SN","qualifierName":"statistics & numerical data"}]}},{"majorTopic_YN":"Y","descriptorName":"Patient Admission","meshQualifierList":{"meshQualifier":[{"majorTopic_YN":"N","abbreviation":"SN","qualifierName":"statistics & numerical data"}]}},{"majorTopic_YN":"N","descriptorName":"Prospective Studies"},{"majorTopic_YN":"Y","descriptorName":"Artificial Intelligence"},{"majorTopic_YN":"Y","descriptorName":"Emergency Service, Hospital","meshQualifierList":{"meshQualifier":[{"majorTopic_YN":"N","abbreviation":"SN","qualifierName":"statistics & numerical data"}]}},{"majorTopic_YN":"N","descriptorName":"Emergency Room Visits"},{"majorTopic_YN":"N","descriptorName":"Prediction Algorithms"}]},"subsetList":{"subset":[{"code":"IM","name":"Index Medicus"}]},"fullTextUrlList":{"fullTextUrl":[{"availability":"Subscription required","availabilityCode":"S","documentStyle":"doi","site":"DOI","url":"https://doi.org/10.1038/s41467-026-72960-1"},{"availability":"Open access","availabilityCode":"OA","documentStyle":"html","site":"Europe_PMC","url":"https://europepmc.org/articles/PMC13377025"},{"availability":"Open access","availabilityCode":"OA","documentStyle":"pdf","site":"Europe_PMC","url":"https://europepmc.org/articles/PMC13377025?pdf=render"}]},"isOpenAccess":"Y","inEPMC":"Y","inPMC":"Y","citedByCount":0,"hasReferences":"Y","hasTextMinedTerms":"Y","hasDbCrossReferences":"N","hasPDF":"Y","hasBook":"N","hasSuppl":"Y","hasLabsLinks":"Y","hasData":"Y","license":"cc by-nc-nd","authMan":"N","epmcAuthMan":"N","nihAuthMan":"N","hasTMAccessionNumbers":"Y","tmAccessionTypeList":{"accessionType":["nct"]},"dateOfCompletion":"2026-07-17","dateOfCreation":"2026-05-12","dateOfRevision":"2026-08-13","electronicPublicationDate":"2026-05-12","firstPublicationDate":"2026-05-12","firstIndexDate":"2026-05-15","fullTextReceivedDate":"2026-09-03","hasEvaluations":"N"},{"id":"41915716","source":"MED","pmid":"41915716","pmcid":"PMC13037975","fullTextIdList":{"fullTextId":["PMC13037975"]},"doi":"10.1371/journal.pmed.1004770","title":"Machine learning assisted differentiation of low acuity patients at dispatch: The MADLAD randomized controlled trial.","authorString":"Spangler DN, Morelli S, Smekal D, Edmark L, Blomberg H.","authorList":{"author":[{"fullName":"Spangler DN","lastName":"Spangler","firstName":"Douglas Nils","initials":"DN","authorId":{"type":"ORCID","value":"0000-0001-6775-5051"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Surgical Sciences, Uppsala University, Uppsala, Sweden."}]}},{"fullName":"Morelli S","lastName":"Morelli","firstName":"Simon","initials":"S","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Centre for Clinical Research, Västmanland Hospital, Västerås, Sweden."}]}},{"fullName":"Smekal D","lastName":"Smekal","firstName":"David","initials":"D","authorId":{"type":"ORCID","value":"0000-0002-3563-6450"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Surgical Sciences, Uppsala University, Uppsala, Sweden."}]}},{"fullName":"Edmark L","lastName":"Edmark","firstName":"Lennart","initials":"L","authorId":{"type":"ORCID","value":"0000-0002-7654-3285"},"authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Surgical Sciences, Uppsala University, Uppsala, Sweden."},{"affiliation":"Centre for Clinical Research, Västmanland Hospital, Västerås, Sweden."}]}},{"fullName":"Blomberg H","lastName":"Blomberg","firstName":"Hans","initials":"H","authorAffiliationDetailsList":{"authorAffiliation":[{"affiliation":"Department of Surgical Sciences, Uppsala University, Uppsala, Sweden."}]}}]},"authorIdList":{"authorId":[{"type":"ORCID","value":"0000-0001-6775-5051"},{"type":"ORCID","value":"0000-0002-3563-6450"},{"type":"ORCID","value":"0000-0002-7654-3285"}]},"dataLinksTagsList":{"dataLinkstag":["altmetrics","supporting_data"]},"journalInfo":{"issue":"3","volume":"23","journalIssueId":4137052,"dateOfPublication":"2026 Mar","monthOfPublication":3,"yearOfPublication":2026,"journal":{"medlineAbbreviation":"PLoS Med","title":"PLoS medicine","essn":"1549-1676","issn":"1549-1277","nlmid":"101231360","isoabbreviation":"PLoS Med"},"printPublicationDate":"2026-03-01"},"pubYear":"2026","pageInfo":"e1004770","abstractText":"<h4>Background</h4>Resource Constrained Situations (RCS) at Emergency Medical Dispatch centers where there are more patients requiring an ambulance than there are available ambulances are common. Machine Learning (ML) techniques offer a promising but largely untested approach to assessing relative risks among these patients. The study aims to establish whether the provision of ML-based risk scores predicting patient outcomes improves the ability of dispatchers to identify patients at high risk for deterioration in RCS and dispatch the first available ambulance to them.<h4>Methods and findings</h4>We performed a parallel-group, randomized trial of adult patients assessed by a dispatch nurse at two study sites in Sweden as requiring a low-priority ambulance response in RCS. Patients were randomized 1:1 to be prioritized with the aid of an ML-based risk assessment tool, or per current clinical practice. The primary outcome was defined in terms of whether the first available ambulance was sent to the patient with the highest National Early Warning Score (NEWS 2) based on subsequently collected vital signs. A total of 1,245 RCS were included in the study. In the intervention arm, 68.3% of RCS were assessed correctly per the primary outcome versus 62.5% in the control group, corresponding to an odds ratio of 1.28 (95% CI [1.00, 1.63], p = 0.047). This study was limited to only patients determined to require a low-priority ambulance response in two Swedish regions, and was underpowered for the primary outcome due to a smaller than expected sample size.<h4>Conclusion</h4>This study suggests that clinical ML-based decision support tools may have the ability to influence care provider decisions and improve their capacity to rapidly differentiate between high- and low-risk patients at dispatch. Further research should establish the suitability of these tools in larger cohorts, for patients with both higher- and lower-levels of priority, and in other settings. The trial was registered at ClinicalTrials.gov (NCT04757194).","affiliation":"Department of Surgical Sciences, Uppsala University, Uppsala, Sweden.","publicationStatus":"epublish","language":"eng","pubModel":"Electronic-eCollection","pubTypeList":{"pubType":["research-article","Multicenter Study","Randomized Controlled Trial","Journal Article"]},"grantsList":{"grant":[{"agency":"VINNOVA","grantId":"2017-04652","orderIn":0}]},"meshHeadingList":{"meshHeading":[{"majorTopic_YN":"N","descriptorName":"Humans"},{"majorTopic_YN":"N","descriptorName":"Risk Assessment"},{"majorTopic_YN":"N","descriptorName":"Ambulances"},{"majorTopic_YN":"N","descriptorName":"Aged"},{"majorTopic_YN":"N","descriptorName":"Middle Aged"},{"majorTopic_YN":"N","descriptorName":"Triage"},{"majorTopic_YN":"N","descriptorName":"Sweden"},{"majorTopic_YN":"N","descriptorName":"Female"},{"majorTopic_YN":"N","descriptorName":"Male"},{"majorTopic_YN":"Y","descriptorName":"Patient Acuity"},{"majorTopic_YN":"Y","descriptorName":"Machine Learning"},{"majorTopic_YN":"Y","descriptorName":"Emergency Medical Dispatch","meshQualifierList":{"meshQualifier":[{"majorTopic_YN":"N","abbreviation":"MT","qualifierName":"methods"}]}},{"majorTopic_YN":"N","descriptorName":"Predictive Learning Models"}]},"subsetList":{"subset":[{"code":"IM","name":"Index Medicus"}]},"fullTextUrlList":{"fullTextUrl":[{"availability":"Subscription required","availabilityCode":"S","documentStyle":"doi","site":"DOI","url":"https://doi.org/10.1371/journal.pmed.1004770"},{"availability":"Open access","availabilityCode":"OA","documentStyle":"html","site":"Europe_PMC","url":"https://europepmc.org/articles/PMC13037975"},{"availability":"Open access","availabilityCode":"OA","documentStyle":"pdf","site":"Europe_PMC","url":"https://europepmc.org/articles/PMC13037975?pdf=render"}]},"isOpenAccess":"Y","inEPMC":"Y","inPMC":"Y","citedByCount":1,"hasReferences":"Y","hasTextMinedTerms":"Y","hasDbCrossReferences":"N","hasPDF":"Y","hasBook":"N","hasSuppl":"Y","hasLabsLinks":"Y","hasData":"Y","license":"cc by","authMan":"N","epmcAuthMan":"N","nihAuthMan":"N","hasTMAccessionNumbers":"Y","tmAccessionTypeList":{"accessionType":["nct","doi"]},"dateOfCompletion":"2026-07-14","dateOfCreation":"2026-03-31","dateOfRevision":"2026-08-14","electronicPublicationDate":"2026-03-31","firstPublicationDate":"2026-03-31","firstIndexDate":"2026-03-31","fullTextReceivedDate":"2026-04-01","hasEvaluations":"N"}]}}