{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T20:26:08Z","timestamp":1787343968992,"version":"build-2736575974"},"reference-count":40,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Acute chest pain (ACP) is one of the most common chief complaints in the emergency department (ED), accounting for approximately 8% of all ED visits. However, among patients presenting with chest pain suggestive of cardiac origin, fewer than 10% are ultimately diagnosed with acute coronary syndrome (ACS).<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n AI models were developed to support clinical decision-making for triage level-2 ED patients presenting with ACP. These models aim to accurately detect ACS and reliably identify low-risk patients based on a single high-sensitivity cardiac troponin T test result. By integrating this AI-assisted strategy into clinical workflows, we aim to reduce ED length of stay, alleviate crowding, and improve health care efficiency while maintaining high safety standards.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n We conducted a retrospective study using single-center data from a tertiary teaching hospital between January 2016 and December 2022. Models based on artificial neural networks (ANNs) were trained to classify patients with ACP at triage level 2 into 3 clinical classes: critical patients with ACS (subgroup GA), critical patients without ACS (subgroup GB1), and low-risk patients (subgroup GB2). The models were trained and internally validated via a 5-fold cross-validation protocol using data from 2016 to 2020. Model performance was then evaluated on the hold-out testing data (2021\u20102022) using area under the receiver operating characteristic curve, area under the precision-recall curve, and subgroup-specific metrics. Feature selection methods and the Shapley Additive Explanations value analysis were applied to identify features that drive the model\u2019s predictive power. The study was approved by the institutional review board of the National Cheng Kung University Hospital, Tainan, Taiwan (A-ER-111\u2010199).<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n After excluding ED visits with missing triage data, incomplete medical histories, or unsuitable dispositions, 17,935 visits were included, with 1209 GA, 3873 GB1, and 12,853 GB2. Rendering prediction based on 24 feature variables (2 demographics, 6 vital signs, 4 blood test results, and 12 medical history), all ANN models demonstrated strong testing AUROC (95% CI) performance of 0.942 (0.920\u20100.965) for GA classification, 0.824 (0.808\u20100.841) for GB1, and 0.893 (0.884\u20100.902) for GB2. Regarding multiclassification, ANN-S3 achieved balanced performance, with ACS sensitivity of 0.941 (95% CI 0.908\u20100.973) and low-risk positive predictive value and sensitivity of 0.911 (95% CI 0.9\u20100.921) and 0.837 (95% CI 0.824\u20100.85), respectively. A sensitivity-prioritized variant, ANN-S3-L, increased ACS sensitivity to 0.966 (95% CI 0.94\u20100.991) and negative predictive value to 0.998 (95% CI 0.996\u20100.999), but at the cost of lower specificity and reduced low-risk sensitivity, indicating a safety-efficiency trade-off.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n These findings suggest that ANN-based classifiers can effectively support clinical risk stratification and disposition decision-making in ACP care for patients presenting \u22653 hours after symptom onset. However, because even the sensitivity-prioritized ANN-S3-L variant falls short of the stringent sensitivity threshold (>0.99) typically required for a standalone ED rule-out tool, this system should be interpreted as a clinical decision-support aid rather than an independent rule-out strategy.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/83099","type":"journal-article","created":{"date-parts":[[2026,7,6]],"date-time":"2026-07-06T17:50:07Z","timestamp":1783360207000},"page":"e83099-e83099","source":"Crossref","is-referenced-by-count":0,"title":["Patient Stratification for Improving Acute Chest Pain Management and Mitigating Emergency Department Crowding: Machine Learning Model Development and Validation"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7465-5742","authenticated-orcid":false,"given":"Chih-Chia","family":"Hsieh","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5636-8985","authenticated-orcid":false,"given":"Shao-Chung","family":"Chu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8638-2235","authenticated-orcid":false,"given":"Jung-Ting","family":"Lee","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0042-0131","authenticated-orcid":false,"given":"Chih-Hao","family":"Lin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0838-7751","authenticated-orcid":false,"given":"Chung-Yao","family":"Kao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,8,21]]},"reference":[{"issue":"23","key":"R1","doi-asserted-by":"publisher","first-page":"2333","DOI":"10.1016\/j.jacc.2022.03.380","article-title":"Care models for acute chest pain that improve outcomes and efficiency: JACC state-of-the-art review","volume":"79","author":"Dawson","journal-title":"J Am Coll Cardiol"},{"issue":"9","key":"R2","doi-asserted-by":"publisher","first-page":"858","DOI":"10.1056\/NEJMoa0900428","article-title":"Early diagnosis of myocardial infarction with sensitive cardiac troponin assays","volume":"361","author":"Reichlin","journal-title":"N Engl J Med"},{"issue":"11","key":"R3","doi-asserted-by":"publisher","first-page":"1109","DOI":"10.1197\/j.aem.2006.07.004","article-title":"Developing models for patient flow and daily surge capacity research","volume":"13","author":"Asplin","journal-title":"Acad Emerg Med"},{"key":"R4","doi-asserted-by":"publisher","DOI":"10.1016\/j.jhealeco.2024.102881","article-title":"Free-for-all: does crowding impact outcomes because hospital emergency departments do not prioritise effectively?","volume":"95","author":"Francetic","journal-title":"J Health Econ"},{"issue":"8","key":"R5","doi-asserted-by":"publisher","first-page":"654","DOI":"10.1097\/JOM.0000000000003126","article-title":"Perceived work demands and emergency department crowding as predictors of objective stress among emergency physicians:a shift-level approach","volume":"66","author":"Britt","journal-title":"J Occup Environ Med"},{"issue":"8","key":"R6","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0252612","article-title":"Applications of machine learning to undifferentiated chest pain in the emergency department: a systematic review","volume":"16","author":"Stewart","journal-title":"PLoS One"},{"issue":"9","key":"R7","doi-asserted-by":"publisher","first-page":"579","DOI":"10.1136\/emermed-2024-214754","article-title":"Improving care with the first measurement of high-sensitivity troponin T (ICare-FirsT) to enable early rule out and reduce length of stay: a diagnostic and observational study","volume":"42","author":"Pickering","journal-title":"Emerg Med J"},{"issue":"23","key":"R8","doi-asserted-by":"publisher","first-page":"2214","DOI":"10.1161\/CIRCULATIONAHA.120.052380","article-title":"High-sensitivity cardiac troponin on presentation to rule out myocardial infarction: a stepped-wedge cluster randomized controlled trial","volume":"143","author":"Anand","journal-title":"Circulation"},{"issue":"3","key":"R9","doi-asserted-by":"publisher","first-page":"304","DOI":"10.1016\/j.cjca.2022.12.028","article-title":"Risk scores for clinical risk stratification of emergency department patients with chest pain but no acute myocardial infarction: a systematic review","volume":"39","author":"O\u2019Rielly","journal-title":"Can J Cardiol"},{"key":"R10","doi-asserted-by":"publisher","DOI":"10.2196\/48862","article-title":"Interpretable deep learning system for identifying critical patients through the prediction of triage level, hospitalization, and length of stay: prospective study","volume":"12","author":"Lin","journal-title":"JMIR Med Inform"},{"issue":"15","key":"R11","doi-asserted-by":"publisher","first-page":"8449","DOI":"10.3390\/app15158449","article-title":"Development and clinical interpretation of an explainable AI model for predicting patient pathways in the emergency department: a retrospective study","volume":"15","author":"Arnaud","journal-title":"Appl Sci"},{"issue":"11","key":"R12","doi-asserted-by":"publisher","first-page":"828","DOI":"10.1016\/S0929-6646(10)60128-3","article-title":"Comparison between Canadian triage and acuity scale and Taiwan triage system in emergency departments","volume":"109","author":"Ng","journal-title":"J Formos Med Assoc"},{"key":"R13","unstructured":"Good machine learning practice for medical device development: guiding principles. US Food and Drug Administration. 2021. URL: https:\/\/www.fda.gov\/medical-devices\/software-medical-device-samd\/good-machine-learning-practice-medical-device-development-guiding-principles [Accessed 04-08-2027]"},{"issue":"18","key":"R14","doi-asserted-by":"publisher","first-page":"2231","DOI":"10.1016\/j.jacc.2018.08.1038","article-title":"Fourth universal definition of myocardial infarction (2018)","volume":"72","author":"Thygesen","journal-title":"J Am Coll Cardiol"},{"issue":"7","key":"R15","doi-asserted-by":"publisher","first-page":"571","DOI":"10.1136\/heartjnl-2020-316891","article-title":"Trends in cardiovascular outcomes after acute coronary syndrome in New Zealand 2006\u20132016","volume":"107","author":"Wang","journal-title":"Heart"},{"key":"R16","unstructured":"Goodfellow I Bengio Y Courville A . Deep Learning. MIT Press; 2016. ISBN: 9780262035613"},{"key":"R17","doi-asserted-by":"publisher","DOI":"10.1016\/j.ajem.2024.06.015","article-title":"Patient stratification based on the risk of severe illness in emergency departments through collaborative machine learning models","volume":"82","author":"Chen","journal-title":"Am J Emerg Med"},{"issue":"1","key":"R18","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1007\/BF00993106","article-title":"Selecting a classification method by cross-validation","volume":"13","author":"Schaffer","journal-title":"Mach Learn"},{"key":"R19","unstructured":"Ioffe S Szegedy C . Batch normalization: accelerating deep network training by reducing internal covariate shift. Presented at: ICML\u201915: Proceedings of the 32nd International Conference on Machine Learning; Jul 6-11, 2015; Lille, France. URL: https:\/\/static.googleusercontent.com\/media\/research.google.com\/en\/\/pubs\/archive\/43442.pdf [Accessed 16-07-2026]"},{"key":"R20","unstructured":"Kingma DP Ba J . Adam: a method for stochastic optimization. arXiv. Preprint posted online on Dec 22, 2014. [doi: 10.48550\/arXiv.1412.6980]"},{"key":"R21","doi-asserted-by":"crossref","unstructured":"Zadrozny B Elkan C . Transforming classifier scores into accurate multiclass probability estimates. Presented at: KDD \u201902: Proceedings of the Eighth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; Jul 23-26, 2002; Edmonton, Alberta, Canada. [doi: 10.1145\/775047.775151]","DOI":"10.1145\/775047.775151"},{"issue":"3","key":"R22","doi-asserted-by":"publisher","first-page":"419","DOI":"10.1002\/bimj.200710415","article-title":"Youden Index and optimal cut\u2010point estimated from observations affected by a lower limit of detection","volume":"50","author":"Ruopp","journal-title":"Biometrical J"},{"issue":"19","key":"R23","doi-asserted-by":"publisher","first-page":"1737","DOI":"10.1056\/NEJMoa2102953","article-title":"New creatinine- and cystatin C-based equations to estimate GFR without race","volume":"385","author":"Inker","journal-title":"N Engl J Med"},{"key":"R24","doi-asserted-by":"crossref","unstructured":"Breiman L et al. Classification and Regression Trees. Routledge; 2017. ISBN: 9781315139470","DOI":"10.1201\/9781315139470"},{"key":"R25","doi-asserted-by":"crossref","unstructured":"Shapley LS . A Value for N-Person Games. The RAND Corporation; 1953.","DOI":"10.1515\/9781400881970-018"},{"issue":"42","key":"R26","doi-asserted-by":"publisher","first-page":"3780","DOI":"10.1093\/eurheartj\/ehy514","article-title":"Impact of age on the performance of the ESC 0\/1h-algorithms for early diagnosis of myocardial infarction","volume":"39","author":"Boeddinghaus","journal-title":"Eur Heart J"},{"issue":"15","key":"R27","doi-asserted-by":"publisher","DOI":"10.3390\/jcm13154416","article-title":"Management of acute coronary syndromes in older people: comprehensive review and multidisciplinary practice-based recommendations","volume":"13","author":"Narendren","journal-title":"J Clin Med"},{"issue":"3","key":"R28","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1016\/j.annemergmed.2010.04.015","article-title":"Older patients in the emergency department: a review","volume":"56","author":"Samaras","journal-title":"Ann Emerg Med"},{"issue":"14","key":"R29","doi-asserted-by":"publisher","first-page":"1289","DOI":"10.1093\/eurheartj\/ehaa575","article-title":"2020 ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: the task force for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation of the European Society of Cardiology (ESC)","volume":"42","author":"Group, E.S.D","journal-title":"Eur Heart J"},{"issue":"21","key":"R30","doi-asserted-by":"publisher","DOI":"10.1161\/JAHA.124.036307","article-title":"Effectiveness and safety of the ESC-TROP (European Society of Cardiology 0h\/1h Troponin Rule-Out Protocol) trial","volume":"13","author":"Mokhtari","journal-title":"J Am Heart Assoc"},{"issue":"8","key":"R31","doi-asserted-by":"publisher","first-page":"1135","DOI":"10.1111\/jch.13622","article-title":"Hypertension and patients with acute coronary syndrome: putting blood pressure levels into perspective","volume":"21","author":"Konstantinou","journal-title":"J of Clinical Hypertension"},{"key":"R32","doi-asserted-by":"publisher","DOI":"10.2147\/CIA.S238930","article-title":"Age-dependent disparities in the prevalence of single and clustering cardiovascular risk factors: a cross-sectional cohort study in middle-aged and older adults","volume":"15","author":"Macek","journal-title":"Clin Interv Aging"},{"issue":"9","key":"R33","doi-asserted-by":"publisher","DOI":"10.1161\/JAHA.121.022198","article-title":"Event rates and risk factors for recurrent cardiovascular events and mortality in a contemporary post acute coronary syndrome population representing 239 234 patients during 2005 to 2018 in the United States","volume":"11","author":"Steen","journal-title":"J Am Heart Assoc"},{"issue":"11","key":"R34","doi-asserted-by":"publisher","first-page":"1157","DOI":"10.1161\/CIRCULATIONAHA.120.050686","article-title":"Cardiovascular disease in chronic kidney disease: pathophysiological insights and therapeutic options","volume":"143","author":"Jankowski","journal-title":"Circulation"},{"key":"R35","doi-asserted-by":"publisher","DOI":"10.2147\/JBM.S457371","article-title":"Haematological indices in acute coronary syndrome patients in Ethiopia: a comparative cross-sectional study","volume":"15","author":"Tadesse","journal-title":"J Blood Med"},{"issue":"1","key":"R36","first-page":"35","volume":"83","author":"de Abreu","journal-title":"MEDICINA (B Aires)"},{"issue":"4","key":"R37","doi-asserted-by":"publisher","first-page":"577","DOI":"10.1016\/j.cjca.2019.09.013","article-title":"Artificial intelligence for diagnosis of acute coronary syndromes: a meta-analysis of machine learning approaches","volume":"36","author":"Iannattone","journal-title":"Can J Cardiol"},{"issue":"1","key":"R38","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.jacc.2023.04.040","article-title":"Single troponin measurement to rule out myocardial infarction: JACC review topic of the week","volume":"82","author":"Jaffe","journal-title":"J Am Coll Cardiol"},{"issue":"12","key":"R39","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.51267","article-title":"Decision fatigue in emergency medicine: an exploration of its validity","volume":"15","author":"Yadav","journal-title":"Cureus"},{"issue":"4","key":"R40","doi-asserted-by":"publisher","first-page":"424","DOI":"10.1016\/j.annemergmed.2025.10.014","article-title":"The Limit of Detection in the Emergency Department Trial (LEGEND): a stepped-wedge cluster randomized trial to rule out acute myocardial infarction and reduce hospital length of stay for patients presenting to the emergency department","volume":"87","author":"Greenslade","journal-title":"Ann Emerg Med"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","deposited":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T19:55:32Z","timestamp":1787342132000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e83099"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,21]]},"references-count":40,"URL":"https:\/\/doi.org\/10.2196\/83099","relation":{},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,21]]},"article-number":"v14i23e83099"}}