{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T15:50:19Z","timestamp":1788277819202,"version":"build-2803163510"},"reference-count":31,"publisher":"Public Library of Science (PLoS)","issue":"12","license":[{"start":{"date-parts":[[2024,12,27]],"date-time":"2024-12-27T00:00:00Z","timestamp":1735257600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004359","name":"Vetenskapsr\u00e5det","doi-asserted-by":"publisher","award":["2019-00198"],"award-info":[{"award-number":["2019-00198"]}],"id":[{"id":"10.13039\/501100004359","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100009963","name":"Forskning och Utveckling, Halland","doi-asserted-by":"publisher","award":["979314"],"award-info":[{"award-number":["979314"]}],"id":[{"id":"10.13039\/501100009963","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Sparbanksstiftelsen","award":["980763"],"award-info":[{"award-number":["980763"]}]},{"name":"Stiftelsen Landsh\u00f6vding Per Westlings minnesfond","award":["RMh2020-0007"],"award-info":[{"award-number":["RMh2020-0007"]}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS ONE"],"abstract":"<jats:sec id=\"sec001\">\n<jats:title>Background<\/jats:title>\n<jats:p>Dyspnoea is one of the emergency department\u2019s (ED) most common and deadly chief complaints, but frequently misdiagnosed and mistreated. We aimed to design a diagnostic decision support which classifies dyspnoeic ED visits into acute heart failure (AHF), exacerbation of chronic obstructive pulmonary disease (eCOPD), pneumonia and \u201cother diagnoses\u201d by using deep learning and complete, unselected data from an entire regional health care system.<\/jats:p>\n<\/jats:sec>\n<jats:sec id=\"sec002\">\n<jats:title>Methods<\/jats:title>\n<jats:p>In this cross-sectional study, we included all dyspnoeic ED visits of patients \u2265 18 years of age at the two EDs in the region of Halland, Sweden, 07\/01\/2017\u201312\/31\/2019. Data from the complete regional health care system within five years prior to the ED visit were analysed. Gold standard diagnoses were defined as the subsequent in-hospital or ED discharge notes, and a subsample was manually reviewed by emergency medicine experts. A novel deep learning model, <jats:italic>the clinical attention-based recurrent encoder network<\/jats:italic> (CareNet), was developed. Cohort performance was compared to a simpler CatBoost model. A list of all variables and their importance for diagnosis was created. For each unique patient visit, the model selected the most important variables, analysed them and presented them to the clinician interpretably by taking event time and clinical context into account. AUROC, sensitivity and specificity were compared.<\/jats:p>\n<\/jats:sec>\n<jats:sec id=\"sec003\">\n<jats:title>Findings<\/jats:title>\n<jats:p>The most prevalent diagnoses among the 10,315 dyspnoeic ED visits were AHF (15.5%), eCOPD (14.0%) and pneumonia (13.3%). Median number of unique events, i.e., registered clinical data with time stamps, per ED visit was 1,095 (IQR 459\u20132,310). CareNet median AUROC was 87.0%, substantially higher than the CatBoost model\u00b4s (81.4%). CareNet median sensitivity for AHF, eCOPD, and pneumonia was 74.5%, 92.6%, and 54.1%, respectively, with a specificity set above 75.0, slightly inferior to that of the CatBoost baseline model. The model assembled a list of 1,596 variables by importance for diagnosis, on top were prior diagnoses of heart failure or COPD, daily smoking, atrial fibrillation\/flutter, life management difficulties and maternity care. Each patient visit received their own unique attention plot, graphically displaying important clinical events for the diagnosis.<\/jats:p>\n<\/jats:sec>\n<jats:sec id=\"sec004\">\n<jats:title>Interpretation<\/jats:title>\n<jats:p>We designed a novel interpretable deep learning model for diagnosis in emergency department dyspnoea patients by analysing unselected data from a complete regional health care system.<\/jats:p>\n<\/jats:sec>","DOI":"10.1371\/journal.pone.0311081","type":"journal-article","created":{"date-parts":[[2024,12,27]],"date-time":"2024-12-27T18:26:56Z","timestamp":1735324016000},"page":"e0311081","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":7,"title":["A novel interpretable deep learning model for diagnosis in emergency department dyspnoea patients based on complete data from an entire health care system"],"prefix":"10.1371","volume":"19","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1106-1701","authenticated-orcid":true,"given":"Ellen T.","family":"Heyman","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5688-0156","authenticated-orcid":true,"given":"Awais","family":"Ashfaq","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ulf","family":"Ekelund","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mattias","family":"Ohlsson","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1883-2000","authenticated-orcid":true,"given":"Jonas","family":"Bj\u00f6rk","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3142-4119","authenticated-orcid":true,"given":"Ardavan M.","family":"Khoshnood","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4068-0341","authenticated-orcid":true,"given":"Markus","family":"Lingman","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2024,12,27]]},"reference":[{"issue":"1","key":"pone.0311081.ref001","doi-asserted-by":"crossref","first-page":"59","DOI":"10.1186\/s13049-021-00874-6","article-title":"Which symptoms pose the highest risk in patients calling for an ambulance? A population-based cohort study from Denmark","volume":"29","author":"S Ibsen","year":"2021","journal-title":"Scand J Trauma Resusc Emerg Med"},{"key":"pone.0311081.ref002","first-page":"4031684","volume":"2022","author":"E Jemt","year":"2022","journal-title":"Outcomes in Emergency Department Patients with Dyspnea versus Chest Pain: A Retrospective Consecutive Cohort Study. Emerg Med Int"},{"issue":"6","key":"pone.0311081.ref003","doi-asserted-by":"crossref","first-page":"855","DOI":"10.5811\/westjem.2022.9.56332","article-title":"Chief Complaints, Underlying Diagnoses, and Mortality in Adult, Non-trauma Emergency Department Visits: A Population-based, Multicenter Cohort Study.","volume":"23","author":"MD Arvig","year":"2022","journal-title":"West J Emerg Med"},{"issue":"3","key":"pone.0311081.ref004","doi-asserted-by":"crossref","first-page":"175","DOI":"10.1097\/MEJ.0b013e3283629c18","article-title":"The acute sick and injured patients: an overview of the emergency department patient population at a Norwegian University Hospital Emergency Department.","volume":"21","author":"NM Langlo","year":"2014","journal-title":"Eur J Emerg Med"},{"issue":"3","key":"pone.0311081.ref005","doi-asserted-by":"crossref","first-page":"R82","DOI":"10.1186\/cc4926","article-title":"Acute respiratory failure in the elderly: etiology, emergency diagnosis and prognosis.","volume":"10","author":"P Ray","year":"2006","journal-title":"Critical care (London, England)."},{"issue":"1","key":"pone.0311081.ref006","doi-asserted-by":"crossref","first-page":"113","DOI":"10.1186\/s13049-016-0305-5","article-title":"Epidemiology, prehospital care and outcomes of patients arriving by ambulance with dyspnoea: an observational study","volume":"24","author":"AM Kelly","year":"2016","journal-title":"Scand J Trauma Resusc Emerg Med"},{"issue":"5","key":"pone.0311081.ref007","doi-asserted-by":"crossref","first-page":"345","DOI":"10.1097\/MEJ.0000000000000571","article-title":"Epidemiology of patients presenting with dyspnea to emergency departments in Europe and the Asia-Pacific region","volume":"26","author":"S Laribi","year":"2019","journal-title":"European journal of emergency medicine"},{"issue":"8","key":"pone.0311081.ref008","doi-asserted-by":"crossref","first-page":"e0271982","DOI":"10.1371\/journal.pone.0271982","article-title":"Shedding light into the black box of out-of-hospital respiratory distress-A retrospective cohort analysis of discharge diagnoses, prehospital diagnostic accuracy, and predictors of mortality.","volume":"17","author":"P Sp\u00f6rl","year":"2022","journal-title":"PLoS One"},{"issue":"2","key":"pone.0311081.ref009","article-title":"Change in diagnosis from the emergency department to hospital discharge in dyspnoeic patients.","volume":"69","author":"SH Ovesen","year":"2022","journal-title":"Dan Med J."},{"issue":"2","key":"pone.0311081.ref010","doi-asserted-by":"crossref","first-page":"267","DOI":"10.1111\/acem.13553","article-title":"High Diagnostic Uncertainty and Inaccuracy in Adult Emergency Department Patients With Dyspnea: A National Database Analysis.","volume":"26","author":"KM Hunold","year":"2019","journal-title":"Academic Emergency Medicine."},{"issue":"1","key":"pone.0311081.ref011","doi-asserted-by":"crossref","first-page":"36","DOI":"10.1186\/s13049-019-0617-3","article-title":"Diagnostic accuracy of physician-staffed emergency medical teams: a retrospective observational cohort study of prehospital versus hospital diagnosis in a 10-year interval","volume":"27","author":"JC Schewe","year":"2019","journal-title":"Scand J Trauma Resusc Emerg Med"},{"issue":"2","key":"pone.0311081.ref012","doi-asserted-by":"crossref","first-page":"184","DOI":"10.1111\/acem.14190","article-title":"Machine Learning Versus Usual Care for Diagnostic and Prognostic Prediction in the Emergency Department: A Systematic Review.","volume":"28","author":"H Kareemi","year":"2021","journal-title":"Acad Emerg Med."},{"key":"pone.0311081.ref013","volume-title":"Rapid Emergency Triage Treatment Scale (RETTS\u00a9) online version 2019.","author":"Predicare","year":"2020"},{"key":"pone.0311081.ref014","unstructured":"World Health O. International Classification of Diseases (ICD) ICD-10 2019. Geneva2020."},{"key":"pone.0311081.ref015","doi-asserted-by":"crossref","DOI":"10.1093\/ije\/dyz262","article-title":"Data resource profile: Regional healthcare information platform in Halland, Sweden, a dedicated environment for healthcare research","author":"A Ashfaq","year":"2020","journal-title":"International journal of epidemiology"},{"key":"pone.0311081.ref016","author":"N. Ibm Corp","year":"2022","journal-title":"IBM SPSS statistics for windows"},{"key":"pone.0311081.ref017","doi-asserted-by":"crossref","unstructured":"Yang Z, Yang D, Dyer C, He X, Smola A, Hovy E, editors. Hierarchical attention networks for document classification. Proceedings of the 2016 conference of the North American chapter of the association for computational linguistics: human language technologies; 2016.","DOI":"10.18653\/v1\/N16-1174"},{"key":"pone.0311081.ref018","article-title":"Empirical evaluation of gated recurrent neural networks on sequence modeling.","author":"J Chung","year":"2014","journal-title":"arXiv preprint arXiv:14123555."},{"key":"pone.0311081.ref019","first-page":"26","article-title":"Distributed representations of words and phrases and their compositionality","author":"T Mikolov","year":"2013","journal-title":"Advances in neural information processing systems"},{"key":"pone.0311081.ref020","article-title":"CatBoost: gradient boosting with categorical features support.","author":"AV Dorogush","year":"2018","journal-title":"arXiv preprint arXiv:181011363."},{"issue":"6","key":"pone.0311081.ref021","doi-asserted-by":"crossref","first-page":"357","DOI":"10.1016\/j.cardfail.2017.08.458","article-title":"Early Identification of Patients With Acute Decompensated Heart Failure.","volume":"24","author":"S Blecker","year":"2018","journal-title":"Journal of Cardiac Failure"},{"issue":"6","key":"pone.0311081.ref022","doi-asserted-by":"crossref","first-page":"875","DOI":"10.5811\/westjem.2019.9.43732","article-title":"Diagnosis of Acute Heart Failure in the Emergency Department: An Evidence-Based Review.","volume":"20","author":"B Long","year":"2019","journal-title":"The western journal of emergency medicine"},{"issue":"11","key":"pone.0311081.ref023","doi-asserted-by":"crossref","first-page":"e0188532","DOI":"10.1371\/journal.pone.0188532","article-title":"A machine learning approach to triaging patients with chronic obstructive pulmonary disease.","volume":"12","author":"S Swaminathan","year":"2017","journal-title":"PloS one"},{"issue":"5","key":"pone.0311081.ref024","doi-asserted-by":"crossref","first-page":"642","DOI":"10.1016\/j.jemermed.2017.05.035","article-title":"Emergency Medicine Evaluation of Community-Acquired Pneumonia: History, Examination, Imaging and Laboratory Assessment, and Risk Scores","volume":"53","author":"B Long","year":"2017","journal-title":"Journal of Emergency Medicine"},{"issue":"1","key":"pone.0311081.ref025","doi-asserted-by":"crossref","first-page":"112","DOI":"10.1093\/bioinformatics\/btr597","article-title":"MissForest\u2014non-parametric missing value imputation for mixed-type data","volume":"28","author":"DJ Stekhoven","year":"2011","journal-title":"Bioinformatics"},{"issue":"2","key":"pone.0311081.ref026","doi-asserted-by":"crossref","first-page":"170","DOI":"10.5811\/westjem.2020.10.48091","article-title":"Diagnostic Uncertainty in Dyspneic Patients with Cancer in the Emergency Department.","volume":"22","author":"KM Hunold","year":"2021","journal-title":"West J Emerg Med"},{"issue":"6","key":"pone.0311081.ref027","doi-asserted-by":"crossref","first-page":"741","DOI":"10.1109\/TRPMS.2021.3066428","article-title":"On Interpretability of Artificial Neural Networks: A Survey.","volume":"5","author":"FL Fan","year":"2021","journal-title":"IEEE Transactions on Radiation and Plasma Medical Sciences."},{"issue":"11","key":"pone.0311081.ref028","doi-asserted-by":"crossref","first-page":"4793","DOI":"10.1109\/TNNLS.2020.3027314","article-title":"A survey on explainable artificial intelligence (xai): Toward medical xai.","volume":"32","author":"E Tjoa","year":"2020","journal-title":"IEEE transactions on neural networks and learning systems"},{"key":"pone.0311081.ref029","doi-asserted-by":"crossref","DOI":"10.23970\/AHRQEPCCER258","article-title":"Diagnostic errors in the emergency department: a systematic review.","author":"DE Newman-Toker","year":"2022"},{"issue":"11","key":"pone.0311081.ref030","doi-asserted-by":"crossref","first-page":"e745","DOI":"10.1016\/S2589-7500(21)00208-9","article-title":"The false hope of current approaches to explainable artificial intelligence in health care","volume":"3","author":"M Ghassemi","year":"2021","journal-title":"Lancet Digit Health"},{"issue":"1","key":"pone.0311081.ref031","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1186\/s40560-019-0393-1","article-title":"Artificial neural networks improve and simplify intensive care mortality prognostication: a national cohort study of 217,289 first-time intensive care unit admissions.","volume":"7","author":"G Holmgren","year":"2019","journal-title":"Journal of intensive care."}],"container-title":["PLOS ONE"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0311081","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,12,27]],"date-time":"2024-12-27T18:27:11Z","timestamp":1735324031000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0311081"}},"subtitle":[],"editor":[{"given":"Amirmohammad","family":"Khalaji","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2024,12,27]]},"references-count":31,"journal-issue":{"issue":"12","published-online":{"date-parts":[[2024,12,27]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0311081","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,12,27]]}}}