{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T17:54:24Z","timestamp":1785779664587,"version":"3.56.0"},"reference-count":37,"publisher":"BMJ","funder":[{"name":"National Science and Technology Council (NSTC), Taiwan","award":["NSTC 113-2221-E-A49-167-MY3"],"award-info":[{"award-number":["NSTC 113-2221-E-A49-167-MY3"]}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["BMJ Qual Saf"],"accepted":{"date-parts":[[2026,7,21]]},"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Sepsis and infection are distinct yet overlapping conditions in the intensive care unit (ICU), posing diagnostic and management challenges due to non-specific clinical features and delayed microbiological confirmation. This study aimed to develop and evaluate a real-time dual-model clinical decision support system for early identification of infection and sepsis based on the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) framework.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We conducted a retrospective model development and prospective non-interventional deployment study of adult ICU admissions between 2018 and 2020 at a tertiary-care medical centre. Infection was defined by positive microbiological culture results while sepsis was defined according to Sepsis-3 criteria. Two machine learning models were developed using structured clinical features from an 8-hour feature window to predict infection and sepsis. The framework included an 8-hour lead time and a 1-hour prediction window. Class imbalance was addressed using propensity score matching and Borderline Synthetic Minority Oversampling Technique. External validation was performed using the Medical Information Mart for Intensive Care IV database. Real-time deployment analyses evaluated ICU risk surveillance and cluster-based stratification using combined sepsis and infection probabilities.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The dual-model system demonstrated consistent discrimination across internal, reduced-feature and external validation cohorts, with area under the receiver operating characteristic curves ranging from 0.75 to 0.85. Both models prioritised high sensitivity and negative predictive value to minimise missed cases. During real-time ICU implementation, the decision support dashboard provided interpretable risk estimates at the bedside. Exploratory pre\u2013post comparisons showed numerically lower point estimates across most clinical and resource-utilisation outcomes, without adjustment for confounding factors. Cluster-based analysis further identified a high-risk subgroup with greater healthcare resource utilisation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>This real-time clinical decision support system enables early identification of infection and sepsis, supports timely clinical decision-making and may inform antibiotic and resource use in the ICU.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/bmjqs-2026-020104","type":"journal-article","created":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T17:20:19Z","timestamp":1785777619000},"page":"bmjqs-2026-020104","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":0,"title":["Real-time clinical decision support system for early identification of infection and sepsis in the intensive care unit: a retrospective development and prospective deployment study"],"prefix":"10.1136","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7016-2609","authenticated-orcid":false,"given":"Fang-Ju","family":"Sun","sequence":"first","affiliation":[{"name":"Institute of Biomedical Informatics, National Yang Ming Chiao Tung University, Taipei, Taiwan"},{"name":"Department of Medical Research, MacKay Memorial Hospital, Taipei, Taiwan"},{"name":"MacKay Junior College of Medicine, Nursing, and Management, Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yen-Yu","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Critical Care Medicine, MacKay Memorial Hospital, Taipei, Taiwan"},{"name":"Institute of Biomedical Sciences, MacKay Medical University, New Taipei, Taiwan"},{"name":"School of Medicine, College of Medicine, MacKay Medical University, New Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Li-Kuo","family":"Kuo","sequence":"additional","affiliation":[{"name":"Department of Critical Care Medicine, MacKay Memorial Hospital, Taipei, Taiwan"},{"name":"School of Medicine, College of Medicine, MacKay Medical University, New Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ting-Yu","family":"Hu","sequence":"additional","affiliation":[{"name":"Department of Critical Care Medicine, MacKay Memorial Hospital, Taipei, Taiwan"},{"name":"School of Medicine, College of Medicine, MacKay Medical University, New Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kuang-Hua","family":"Cheng","sequence":"additional","affiliation":[{"name":"Department of Critical Care Medicine, MacKay Memorial Hospital, Taipei, Taiwan"},{"name":"School of Medicine, College of Medicine, MacKay Medical University, New Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hung-Ting","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Information Technology, MacKay Memorial Hospital, Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Po-Jen","family":"Chang","sequence":"additional","affiliation":[{"name":"Department of Information Technology, MacKay Memorial Hospital, Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Min-Ching","family":"Wu","sequence":"additional","affiliation":[{"name":"Department of Finance, MacKay Memorial Hospital, Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hung-I","family":"Yeh","sequence":"additional","affiliation":[{"name":"Department of Medical Research, MacKay Memorial Hospital, Taipei, Taiwan"},{"name":"School of Medicine, College of Medicine, MacKay Medical University, New Taipei, Taiwan"},{"name":"Department of Cardiovascular Medicine\/Cardiovascular Center, MacKay Memorial Hospital, Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kun-Pin","family":"Wu","sequence":"additional","affiliation":[{"name":"Institute of Biomedical Informatics, National Yang Ming Chiao Tung University, Taipei, Taiwan"},{"name":"Ph.D. Program of Interdisciplinary Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2026,8,3]]},"reference":[{"key":"2026080310201351000_bmjqs-2026-020104v1.1","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2016.0287"},{"key":"2026080310201351000_bmjqs-2026-020104v1.2","doi-asserted-by":"publisher","DOI":"10.1097\/ccm.0000000000005337"},{"key":"2026080310201351000_bmjqs-2026-020104v1.3","doi-asserted-by":"publisher","DOI":"10.1097\/CCM.0000000000002936"},{"key":"2026080310201351000_bmjqs-2026-020104v1.4","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1007\/s15010-023-02080-1","article-title":"Evaluation of the clinical relevance of the Biofire\u00a9 FilmArray pneumonia panel among hospitalized patients","volume":"52","author":"S\u00f8gaard","year":"2024","journal-title":"Infection"},{"key":"2026080310201351000_bmjqs-2026-020104v1.5","doi-asserted-by":"crossref","first-page":"104110","DOI":"10.1016\/j.compbiomed.2020.104110","article-title":"SSP: Early prediction of sepsis using fully connected LSTM-CNN model","volume":"128","author":"Rafiei","year":"2021","journal-title":"Comput Biol Med"},{"key":"2026080310201351000_bmjqs-2026-020104v1.6","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmii.2023.05.001"},{"key":"2026080310201351000_bmjqs-2026-020104v1.7","doi-asserted-by":"crossref","first-page":"533","DOI":"10.1016\/j.csbj.2024.07.018","article-title":"A step forward in the diagnosis of urinary tract infections: from machine learning to clinical practice","volume":"24","author":"Flores","year":"2024","journal-title":"Comput Struct Biotechnol J"},{"key":"2026080310201351000_bmjqs-2026-020104v1.8","doi-asserted-by":"crossref","DOI":"10.1186\/s12931-023-02624-x","article-title":"Using time-series chest radiographs and laboratory data by machine learning for identifying pulmonary infection and colonization of Acinetobacter baumannii","volume":"25","author":"Zeng","year":"2024","journal-title":"Respir Res"},{"key":"2026080310201351000_bmjqs-2026-020104v1.9","doi-asserted-by":"crossref","unstructured":"Arik S\u00d6 , Pfister T . TabNet: attentive interpretable tabular learning. proc aaai conf artif intell 2021;35:6679\u201387. doi:10.1609\/aaai.v35i8.16826","DOI":"10.1609\/aaai.v35i8.16826"},{"key":"2026080310201351000_bmjqs-2026-020104v1.10","doi-asserted-by":"crossref","first-page":"107418","DOI":"10.1016\/j.compbiomed.2023.107418","article-title":"Feature augmentation and semi-supervised conditional transfer learning for early detection of sepsis","volume":"165","author":"Dou","year":"2023","journal-title":"Comput Biol Med"},{"key":"2026080310201351000_bmjqs-2026-020104v1.11","doi-asserted-by":"publisher","DOI":"10.1016\/j.ccc.2019.08.002"},{"key":"2026080310201351000_bmjqs-2026-020104v1.12","doi-asserted-by":"publisher","DOI":"10.1007\/s00134-022-06956-y"},{"key":"2026080310201351000_bmjqs-2026-020104v1.13","doi-asserted-by":"publisher","DOI":"10.1016\/0021-9681(87)90171-8"},{"key":"2026080310201351000_bmjqs-2026-020104v1.14","doi-asserted-by":"publisher","DOI":"10.1613\/jair.953"},{"key":"2026080310201351000_bmjqs-2026-020104v1.15","doi-asserted-by":"crossref","DOI":"10.1038\/s41597-022-01899-x","article-title":"MIMIC-IV, a freely accessible electronic health record dataset","volume":"10","author":"Johnson","year":"2023","journal-title":"Sci Data"},{"key":"2026080310201351000_bmjqs-2026-020104v1.16","doi-asserted-by":"publisher","DOI":"10.2307\/2290268"},{"key":"2026080310201351000_bmjqs-2026-020104v1.17","doi-asserted-by":"publisher","DOI":"10.2307\/2531595"},{"key":"2026080310201351000_bmjqs-2026-020104v1.18","doi-asserted-by":"crossref","DOI":"10.3389\/fmed.2024.1510792","article-title":"Harnessing artificial intelligence in sepsis care: advances in early detection, personalized treatment, and real-time monitoring","volume":"11","author":"Li","year":"2024","journal-title":"Front Med (Lausanne)"},{"key":"2026080310201351000_bmjqs-2026-020104v1.19","doi-asserted-by":"crossref","first-page":"293","DOI":"10.1186\/s13054-018-2234-3","article-title":"Cost-effectiveness of procalcitonin testing to guide antibiotic treatment duration in critically ill patients: results from a randomised controlled multicentre trial in the Netherlands","volume":"22","author":"Kip","year":"2018","journal-title":"Crit Care"},{"key":"2026080310201351000_bmjqs-2026-020104v1.20","doi-asserted-by":"crossref","DOI":"10.1186\/s13613-021-00931-4","article-title":"Impact on antimicrobial consumption of procalcitonin-guided antibiotic therapy for pneumonia\/pneumonitis associated with aspiration in comatose mechanically ventilated patients: a multicenter, randomized controlled study","volume":"11","author":"Labro","year":"2021","journal-title":"Ann Intensive Care"},{"key":"2026080310201351000_bmjqs-2026-020104v1.21","doi-asserted-by":"crossref","DOI":"10.1038\/s41598-023-49890-9","article-title":"Deep learning-based prediction of in-hospital mortality for sepsis","volume":"14","author":"Yong","year":"2024","journal-title":"Sci Rep"},{"key":"2026080310201351000_bmjqs-2026-020104v1.22","doi-asserted-by":"crossref","DOI":"10.1136\/bmjopen-2017-017833","article-title":"Multicentre validation of a sepsis prediction algorithm using only vital sign data in the emergency department, general ward and ICU","volume":"8","author":"Mao","year":"2018","journal-title":"BMJ Open"},{"key":"2026080310201351000_bmjqs-2026-020104v1.23","doi-asserted-by":"crossref","DOI":"10.1038\/s41467-021-20910-4","article-title":"Artificial intelligence in sepsis early prediction and diagnosis using unstructured data in healthcare","volume":"12","author":"Goh","year":"2021","journal-title":"Nat Commun"},{"key":"2026080310201351000_bmjqs-2026-020104v1.24","doi-asserted-by":"crossref","first-page":"210","DOI":"10.1097\/CCM.0000000000004145","article-title":"Early Prediction of Sepsis From Clinical Data: The PhysioNet\/Computing in Cardiology Challenge 2019","volume":"48","author":"Reyna","year":"2020","journal-title":"Crit Care Med"},{"key":"2026080310201351000_bmjqs-2026-020104v1.25","doi-asserted-by":"crossref","first-page":"101820","DOI":"10.1016\/j.artmed.2020.101820","article-title":"Early detection of sepsis utilizing deep learning on electronic health record event sequences","volume":"104","author":"Lauritsen","year":"2020","journal-title":"Artif Intell Med"},{"key":"2026080310201351000_bmjqs-2026-020104v1.26","doi-asserted-by":"crossref","first-page":"103957","DOI":"10.1016\/j.compbiomed.2020.103957","article-title":"Automated prediction of sepsis using temporal convolutional network","volume":"127","author":"Kok","year":"2020","journal-title":"Comput Biol Med"},{"key":"2026080310201351000_bmjqs-2026-020104v1.27","doi-asserted-by":"crossref","DOI":"10.1016\/j.eclinm.2023.102124","article-title":"Predicting sepsis using deep learning across international sites: a retrospective development and validation study","volume":"62","author":"Moor","year":"2023","journal-title":"EClinicalMedicine"},{"key":"2026080310201351000_bmjqs-2026-020104v1.28","doi-asserted-by":"crossref","first-page":"e1020","DOI":"10.1097\/CCM.0000000000004556","article-title":"The Development and Validation of a Machine Learning Model to Predict Bacteremia and Fungemia in Hospitalized Patients Using Electronic Health Record Data","volume":"48","author":"Bhavani","year":"2020","journal-title":"Crit Care Med"},{"key":"2026080310201351000_bmjqs-2026-020104v1.29","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcrc.2018.02.010"},{"key":"2026080310201351000_bmjqs-2026-020104v1.30","doi-asserted-by":"crossref","DOI":"10.1371\/journal.pdig.0000261","article-title":"Can the application of machine learning to electronic health records guide antibiotic prescribing decisions for suspected urinary tract infection in the Emergency Department?","volume":"2","author":"Rockenschaub","year":"2023","journal-title":"PLOS Digit Health"},{"key":"2026080310201351000_bmjqs-2026-020104v1.31","doi-asserted-by":"crossref","first-page":"1108","DOI":"10.1093\/jac\/dky514","article-title":"Supervised machine learning for the prediction of infection on admission to hospital: a prospective observational cohort study","volume":"74","author":"Rawson","year":"2019","journal-title":"J Antimicrob Chemother"},{"key":"2026080310201351000_bmjqs-2026-020104v1.32","doi-asserted-by":"crossref","DOI":"10.1001\/jamanetworkopen.2023.29729","article-title":"Sepsis Prediction Model for Determining Sepsis vs SIRS, qSOFA, and SOFA","volume":"6","author":"Schertz","year":"2023","journal-title":"JAMA Netw Open"},{"key":"2026080310201351000_bmjqs-2026-020104v1.33","doi-asserted-by":"crossref","DOI":"10.1186\/s13054-024-04948-6","article-title":"A scoping review of machine learning for sepsis prediction- feature engineering strategies and model performance: a step towards explainability","volume":"28","author":"Bomrah","year":"2024","journal-title":"Crit Care"},{"key":"2026080310201351000_bmjqs-2026-020104v1.34","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.309.6947.102"},{"key":"2026080310201351000_bmjqs-2026-020104v1.35","doi-asserted-by":"crossref","first-page":"503","DOI":"10.3390\/medicina57050503","article-title":"Foundational Statistical Principles in Medical Research: Sensitivity, Specificity, Positive Predictive Value, and Negative Predictive Value","volume":"57","author":"Monaghan","year":"2021","journal-title":"Medicina (Kaunas)"},{"key":"2026080310201351000_bmjqs-2026-020104v1.36","doi-asserted-by":"crossref","DOI":"10.1186\/s12916-019-1466-7","article-title":"Calibration: the Achilles heel of predictive analytics","volume":"17","author":"Van Calster","year":"2019","journal-title":"BMC Med"},{"key":"2026080310201351000_bmjqs-2026-020104v1.37","doi-asserted-by":"publisher","DOI":"10.1515\/mr-2023-0039"}],"container-title":["BMJ Quality &amp; Safety"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjqs-2026-020104","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T17:20:23Z","timestamp":1785777623000},"score":1,"resource":{"primary":{"URL":"https:\/\/qualitysafety.bmj.com\/lookup\/doi\/10.1136\/bmjqs-2026-020104"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,3]]},"references-count":37,"alternative-id":["10.1136\/bmjqs-2026-020104"],"URL":"https:\/\/doi.org\/10.1136\/bmjqs-2026-020104","relation":{},"ISSN":["2044-5415","2044-5423"],"issn-type":[{"value":"2044-5415","type":"print"},{"value":"2044-5423","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,3]]}}}