{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,29]],"date-time":"2026-05-29T14:34:50Z","timestamp":1780065290696,"version":"3.54.0"},"reference-count":25,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2025,12,20]],"date-time":"2025-12-20T00:00:00Z","timestamp":1766188800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,12,29]],"date-time":"2025-12-29T00:00:00Z","timestamp":1766966400000},"content-version":"vor","delay-in-days":9,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["BioData Mining"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Emergency Department (ED) revisits represent a critical issue in emergency medicine. Identifying high-risk revisit cases (revisits with intensive care unit admissions, cardiac arrest, or requiring emergency surgery) is particularly important. While prior studies have explored machine learning models for ED revisit prediction, few deep learning approaches exist, and dynamic features remain underutilized.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We used data from National Taiwan University Hospital (NTUH), incorporating both static (e.g., age, sex, triage) and dynamic (vital signs) features. A preprocessing strategy was developed to handle temporal irregularities. We proposed a hybrid deep learning model combining Temporal Convolutional Network (TCN) and feature tokenizer (FT)-Transformer to integrate static and short-term dynamic information.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>We evaluated our model on NTUH 2016\u20132019 data, achieving the area under the receiver operating characteristic curve (AUROC) of 0.8453 and the area under precision recall curve (AUPRC) of 0.0935 for high-risk revisits (base rate\u2009=\u20090.01), and AUROC of 0.7250 and AUPRC of 0.2005 for general revisits (base rate\u2009=\u20090.042). The model maintained robust performance when validated on 2020\u20132022 data. Compared to the static-only logistic regression baseline, our model improved AUPRC from 0.0288 to 0.0935 and precision from 0.0281 to 0.0428.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Our model significantly outperformed the static-only baseline. It demonstrates the effectiveness of multimodal clinical data fusion in improving ED revisit prediction and supporting clinical decision-making.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s13040-025-00509-x","type":"journal-article","created":{"date-parts":[[2025,12,20]],"date-time":"2025-12-20T16:46:29Z","timestamp":1766249189000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Deep learning to predict emergency department revisit using static and dynamic features (Deep Revisit): development and validation study"],"prefix":"10.1186","volume":"18","author":[{"given":"Su-Yin","family":"Hsu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jhe-Yi","family":"Jhu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jun-Wan","family":"Gao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chien-Hua","family":"Huang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chu-Lin","family":"Tsai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Li-Chen","family":"Fu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2025,12,20]]},"reference":[{"key":"509_CR1","unstructured":"Ministry of health and welfare, Taiwan: 2022 health and welfare report \u2013 emergency department statistics. Accessed: 2025-04-08. 2022. https:\/\/dep.mohw.gov.tw\/dos\/lp-5103-113-xCat-y111.html."},{"key":"509_CR2","doi-asserted-by":"publisher","unstructured":"Sung C-W, Lu T-C, Fang C-C, Lin J-Y, Yeh H-F, Huang C-H, Tsai C-L. Factors associated with a high-risk return visit to the emergency department: a case-crossover study. Eur J Emerg Med. 2021;28(5):394\u2013401. https:\/\/doi.org\/10.1097\/mej.0000000000000851.","DOI":"10.1097\/mej.0000000000000851"},{"key":"509_CR3","doi-asserted-by":"publisher","unstructured":"Si Y, Du J, Li Z, Jiang X, Miller T, Wang F, Jim Zheng W, Roberts K. Deep representation learning of patient data from electronic health records (ehr): a systematic review. J Biomed Inf. 2021;115:103671. https:\/\/doi.org\/10.1016\/j.jbi.2020.103671.","DOI":"10.1016\/j.jbi.2020.103671"},{"key":"509_CR4","doi-asserted-by":"publisher","unstructured":"Pellerin G, Gao K, Kaminsky L. Predicting 72-hour emergency department revisits. Am J Emerg Med. 2018;36(3):420\u201324. https:\/\/doi.org\/10.1016\/j.ajem.2017.08.049.","DOI":"10.1016\/j.ajem.2017.08.049"},{"key":"509_CR5","doi-asserted-by":"publisher","unstructured":"Porto BM. Improving triage performance in emergency departments using machine learning and natural language processing: a systematic review. BMC Emerg Med. 2024;24(1). https:\/\/doi.org\/10.1186\/s12873-024-01135-2.","DOI":"10.1186\/s12873-024-01135-2"},{"key":"509_CR6","doi-asserted-by":"publisher","unstructured":"Kuo K-M, Wu W-S, Chang CS. A meta-analysis of the diagnostic test accuracy of artificial intelligence for predicting emergency department revisits. J Med Syst. 2025;49(1). https:\/\/doi.org\/10.1007\/s10916-025-02210-2.","DOI":"10.1007\/s10916-025-02210-2"},{"key":"509_CR7","doi-asserted-by":"publisher","unstructured":"Lee Y-C, Ng C-J, Hsu C-C, Cheng C-W, Chen S-Y. Machine learning models for predicting unscheduled return visits to an emergency department: a scoping review. BMC Emerg Med. 2024;24(1). https:\/\/doi.org\/10.1186\/s12873-024-00939-6.","DOI":"10.1186\/s12873-024-00939-6"},{"key":"509_CR8","doi-asserted-by":"publisher","unstructured":"Chen T, Guestrin C. Xgboost: a scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. KDD\u201916. New York, NY, USA: ACM; 2016, pp. 785\u201394). https:\/\/doi.org\/10.1145\/2939672.2939785.","DOI":"10.1145\/2939672.2939785"},{"issue":"7","key":"509_CR9","doi-asserted-by":"publisher","first-page":"0218942","DOI":"10.1371\/journal.pone.0218942","volume":"14","author":"Y-W Lin","year":"2019","unstructured":"Lin Y-W, Zhou Y, Faghri F, Shaw MJ, Campbell RH. Analysis and prediction of unplanned intensive care unit readmission using recurrent neural networks with long short-term memory. PLoS One. 2019;14(7):0218942. https:\/\/doi.org\/10.1371\/journal.pone.0218942.","journal-title":"PLoS One"},{"key":"509_CR10","doi-asserted-by":"publisher","unstructured":"Zebin T, Chaussalet TJ. Design and implementation of a deep recurrent model for prediction of readmission in urgent care using electronic health records. 2019 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (CIBCB). Siena, Italy: IEEE; 2019, pp. 1\u20135). doi: https:\/\/doi.org\/10.1109\/CIBCB.2019.8791466.","DOI":"10.1109\/CIBCB.2019.8791466"},{"key":"509_CR11","doi-asserted-by":"publisher","unstructured":"Porto BM, Fogliatto FS. Enhanced forecasting of emergency department patient arrivals using feature engineering approach and machine learning. BMC Med Inf Decis Mak. 2024;24(1). https:\/\/doi.org\/10.1186\/s12911-024-02788-6.","DOI":"10.1186\/s12911-024-02788-6"},{"key":"509_CR12","doi-asserted-by":"publisher","unstructured":"Sung C-W, Ho J, Fan C-Y, Chen C-Y, Chen C-H, Lin S-Y, Chang J-H, Chen J-W, Huang EP-C. Prediction of high-risk emergency department revisits from a machine-learning algorithm: a proof-of-concept study. BMJ Health Care Inf. 2024;31(1):100859. https:\/\/doi.org\/10.1136\/bmjhci-2023-100859.","DOI":"10.1136\/bmjhci-2023-100859"},{"issue":"6","key":"509_CR13","doi-asserted-by":"publisher","first-page":"3702","DOI":"10.1002\/ehf2.14918","volume":"11","author":"M Kim","year":"2024","unstructured":"Kim M, Lee YS, Park Y, Jung A, So H, Park J, Park J, Choi D, Kim S, Park S. Deep learning for predicting rehospitalization in acute heart failure: Model foundation and external validation. ESC Heart Fail. 2024;11(6):3702\u201312. https:\/\/doi.org\/10.1002\/ehf2.14918.","journal-title":"ESC Heart Fail"},{"key":"509_CR14","doi-asserted-by":"publisher","unstructured":"Lopez K, Li H, Lipkin-Moore Z, Kay S, Rajeevan H, Davis JL, Wilson FP, Rochester CL, Gomez JL. Deep learning prediction of hospital readmissions for asthma and copd. Respir Res. 2023;24(1). https:\/\/doi.org\/10.1186\/s12931-023-02628-7.","DOI":"10.1186\/s12931-023-02628-7"},{"key":"509_CR15","doi-asserted-by":"publisher","unstructured":"Ashfaq A, Sant\u2019anna A, Lingman M, Nowaczyk S. Readmission prediction using deep learning on electronic health records. J Biomed Inf. 2019;97:103256. https:\/\/doi.org\/10.1016\/j.jbi.2019.103256.","DOI":"10.1016\/j.jbi.2019.103256"},{"key":"509_CR16","doi-asserted-by":"publisher","unstructured":"Tang S, Tariq A, Dunnmon JA, Sharma U, Elugunti P, Rubin DL, Patel BN, Banerjee I. Predicting 30-day all-cause hospital readmission using multimodal spatiotemporal graph neural networks. IEEE J Biomed Health Inform 2023, 1\u201312) doi: https:\/\/doi.org\/10.1109\/jbhi.2023.3236888.","DOI":"10.1109\/jbhi.2023.3236888"},{"issue":"7","key":"509_CR17","doi-asserted-by":"publisher","first-page":"0254841","DOI":"10.1371\/journal.pone.0254841","volume":"16","author":"BK Iwana","year":"2021","unstructured":"Iwana BK, Uchida S. An empirical survey of data augmentation for time series classification with neural networks. PLoS One. 2021;16(7):0254841. https:\/\/doi.org\/10.1371\/journal.pone.0254841.","journal-title":"PLoS One"},{"issue":"10","key":"509_CR18","doi-asserted-by":"publisher","first-page":"9825","DOI":"10.3934\/mbe.2022457","volume":"19","author":"Q Tang","year":"2022","unstructured":"Tang Q, Cen X, Pan C. Explainable and efficient deep early warning system for cardiac arrest prediction from electronic health records. Math Biosci And Eng. 2022;19(10):9825\u201341. https:\/\/doi.org\/10.3934\/mbe.2022457.","journal-title":"Math Biosci And Eng"},{"key":"509_CR19","doi-asserted-by":"publisher","first-page":"104440","DOI":"10.1016\/j.jbi.2023.104440","volume":"144","author":"M Kazijevs","year":"2023","unstructured":"Kazijevs M, Samad MD. Deep imputation of missing values in time series health data: a review with benchmarking. J Biomed Inf. 2023;144:104440. https:\/\/doi.org\/10.1016\/j.jbi.2023.104440.","journal-title":"J Biomed Inf"},{"key":"509_CR20","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1803.01271","volume-title":"An empirical evaluation of generic Convolutional and recurrent networks for sequence modeling. arXiv","author":"S Bai","year":"2018","unstructured":"Bai S, Kolter JZ, Koltun V. An empirical evaluation of generic Convolutional and recurrent networks for sequence modeling. arXiv. 2018. https:\/\/doi.org\/10.48550\/ARXIV.1803.01271."},{"key":"509_CR21","doi-asserted-by":"publisher","unstructured":"Wang J-Y, Hsu S-Y, Sun J-T, Ko C-H, Huang C-H, Tsai C-L, Fu L-C. Internal and external validation of a deep learning-based early warning system of cardiac arrest with variable-length and irregularly measured time series data. J Healthcare Inf Res. 2025. https:\/\/doi.org\/10.1007\/s41666-025-00188-7.","DOI":"10.1007\/s41666-025-00188-7"},{"key":"509_CR22","doi-asserted-by":"publisher","unstructured":"Namgung M, Lee DH, Bae SJ, Chung HS, Kim K, Lee CA, Kim DH, Kim EC, Lim JY, Han SS, Choi YH. The impact of covid-19 pandemic on revisits to emergency department. Australas Emerg Care. 2023;26(3):221\u201329. https:\/\/doi.org\/10.1016\/j.auec.2023.01.002.","DOI":"10.1016\/j.auec.2023.01.002"},{"key":"509_CR23","doi-asserted-by":"publisher","unstructured":"Hond AAH, Shah VB, Kant IMJ, Van Calster B, Steyerberg EW, Hernandez-Boussard T. Perspectives on validation of clinical predictive algorithms. npj Digit Med 6(1). 2023. https:\/\/doi.org\/10.1038\/s41746-023-00832-9.","DOI":"10.1038\/s41746-023-00832-9"},{"key":"509_CR24","doi-asserted-by":"publisher","unstructured":"Moons KGM, Altman DG, Reitsma JB, Ioannidis JPA, Macaskill P, Steyerberg EW, Vickers AJ, Ransohoff DF, Collins GS. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (tripod): explanation and elaboration. Ann Intern Med. 2015;162(1):1\u201373. https:\/\/doi.org\/10.7326\/m14-0698.","DOI":"10.7326\/m14-0698"},{"key":"509_CR25","unstructured":"Ismail F, A. Benchmarking deep learning interpretability in time series classification. Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. 2020, pp. 2633\u201343."}],"container-title":["BioData Mining"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s13040-025-00509-x","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13040-025-00509-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13040-025-00509-x.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,29]],"date-time":"2025-12-29T20:56:05Z","timestamp":1767041765000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1186\/s13040-025-00509-x"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,20]]},"references-count":25,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,12]]}},"alternative-id":["509"],"URL":"https:\/\/doi.org\/10.1186\/s13040-025-00509-x","relation":{},"ISSN":["1756-0381"],"issn-type":[{"value":"1756-0381","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,20]]},"assertion":[{"value":"22 August 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"12 December 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"20 December 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"This study was approved by the National Taiwan Hospital Institutional Review Board (202207128RINC). All methods were performed in accordance with the relevant guidelines and regulations, including the Declaration of Helsinki.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethics approval and consent to participate"}},{"value":"Not applicable.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent for publication"}},{"value":"The authors declare no competing interests.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"88"}}