{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,30]],"date-time":"2026-09-30T16:54:44Z","timestamp":1790787284037,"version":"4.1.0"},"update-to":[{"DOI":"10.1371\/journal.pmed.1005024","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T00:00:00Z","timestamp":1775088000000}}],"reference-count":40,"publisher":"Public Library of Science (PLoS)","issue":"3","license":[{"start":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T00:00:00Z","timestamp":1774396800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["81730033"],"award-info":[{"award-number":["81730033"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["82171193"],"award-info":[{"award-number":["82171193"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Key Talent Project for Strengthening Health during the 13th Five-Year Plan Period","award":["ZDRCA2016069"],"award-info":[{"award-number":["ZDRCA2016069"]}]},{"DOI":"10.13039\/501100012166","name":"National Key R&D Program of China","doi-asserted-by":"crossref","award":["2018YFC2001901"],"award-info":[{"award-number":["2018YFC2001901"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Jiangsu Provincial Medical Key Discipline","award":["ZDXK202232"],"award-info":[{"award-number":["ZDXK202232"]}]},{"name":"Clinical Trials from the Affiliated Drum Tower Hospital, Medical School of Nanjing University","award":["2021-LCYJ-DBZ-04"],"award-info":[{"award-number":["2021-LCYJ-DBZ-04"]}]},{"name":"Clinical Trials from the Affiliated Drum Tower Hospital, Medical School of Nanjing University","award":["2024-LCYJ-MS-18"],"award-info":[{"award-number":["2024-LCYJ-MS-18"]}]}],"content-domain":{"domain":["www.plosmedicine.org"],"crossmark-restriction":false},"short-container-title":["PLoS Med"],"abstract":"<jats:sec id=\"sec001\">\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>The clinical importance of transient intraoperative hypotension (IOH) remains debated, and existing models often rely on high-resolution waveform data that are not routinely available.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Methods and findings<\/jats:title>\n                    <jats:p>\n                      We developed a Transformer-based deep learning model to predict IOH in real time using continuous vital sign time-series data. The model was trained on 319,699 surgical cases from a tertiary hospital in China (2013\u20132023) and externally validated using an independent dataset from South Korea. Model interpretability was explored through a real-time alert simulation using 10 representative surgical cases from the internal validation cohort, comparing predicted IOH risk trajectories with measured mean arterial pressure (MAP). To assess clinical relevance, a nested cohort study evaluated the association between IOH burden (cumulative MAP\u2009\u226465\/60\/55 mmHg in mmHg\u00b7min) and postoperative acute kidney injury (AKI) and acute kidney disease (AKD). The Transformer model achieved strong prediction performance at 5-, 10-, and 15-min horizons (AUCs 0.904, 0.892, 0.882; recall \u226588.3%). Compared with XGBoost, the Transformer had higher recall (internal 5-min recall 0.891 versus 0.737) and substantially better probability calibration (expected calibration error 0.0083 versus 0.0373). XGBoost showed higher overall accuracy and specificity (internal 5-min specificity 0.913 versus 0.723). External validation confirmed comparable discrimination across models and generalizability, with attenuated calibration differences. In alert simulations, predicted IOH risk closely corresponded to MAP fluctuations. IOH burden was significantly associated with postoperative AKI and AKD (MAP\u2009\u226465 mmHg: OR per 60 mmHg\u00b7min 1.10 (95% CI, [1.02, 1.19];\n                      <jats:italic>p<\/jats:italic>\n                      \u2009=\u20090.012) for AKI; 1.26 (95% CI, [1.19, 1.33];\n                      <jats:italic>p<\/jats:italic>\n                      \u2009&lt;\u20090.001) for AKD). The study is limited by its retrospective design, and prospective, multicenter validation is needed to confirm real-time applicability and generalizability of the model.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>IOH burden is associated with increased risk of postoperative AKI and AKD. The Transformer model prioritizes sensitivity and calibration for short-term IOH prediction, whereas XGBoost emphasizes accuracy and specificity, reflecting different operating characteristics. Prospective, real-time evaluation is needed before clinical implementation.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pmed.1005024","type":"journal-article","created":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T17:40:23Z","timestamp":1774460423000},"page":"e1005024","update-policy":"https:\/\/doi.org\/10.1371\/journal.pmed.corrections_policy","source":"Crossref","is-referenced-by-count":5,"title":["Transformer-based deep learning model for real-time prediction of intraoperative hypotension using dynamic time-series vital signs: A retrospective study"],"prefix":"10.1371","volume":"23","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-2021-5781","authenticated-orcid":true,"given":"Shouqiang","family":"Zhu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weihai","family":"Shi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haichuan","family":"Qian","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaoqiang","family":"Tong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rongliang","family":"Hu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinhua","family":"Bo","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8218-7299","authenticated-orcid":true,"given":"Xiaoping","family":"Gu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,3,25]]},"reference":[{"issue":"5","key":"pmed.1005024.ref001","doi-asserted-by":"crossref","first-page":"1246","DOI":"10.1213\/ANE.0000000000003668","article-title":"Multimodal general anesthesia: theory and practice","volume":"127","author":"EN Brown","year":"2018","journal-title":"Anesth Analg"},{"issue":"6","key":"pmed.1005024.ref002","doi-asserted-by":"crossref","first-page":"811","DOI":"10.1007\/s00134-018-5224-7","article-title":"Perioperative myocardial injury and the contribution of hypotension","volume":"44","author":"DI Sessler","year":"2018","journal-title":"Intensive Care Med"},{"issue":"1","key":"pmed.1005024.ref003","doi-asserted-by":"crossref","first-page":"47","DOI":"10.1097\/ALN.0000000000001432","article-title":"Relationship between intraoperative hypotension, defined by either reduction from baseline or absolute thresholds, and acute kidney and myocardial injury after noncardiac surgery: a retrospective cohort analysis","volume":"126","author":"V Salmasi","year":"2017","journal-title":"Anesthesiology"},{"issue":"3","key":"pmed.1005024.ref004","doi-asserted-by":"crossref","first-page":"507","DOI":"10.1097\/ALN.0b013e3182a10e26","article-title":"Relationship between intraoperative mean arterial pressure and clinical outcomes after noncardiac surgery: toward an empirical definition of hypotension","volume":"119","author":"M Walsh","year":"2013","journal-title":"Anesthesiology"},{"issue":"3","key":"pmed.1005024.ref005","doi-asserted-by":"crossref","first-page":"515","DOI":"10.1097\/ALN.0000000000000765","article-title":"Association of intraoperative hypotension with acute kidney injury after elective noncardiac surgery","volume":"123","author":"LY Sun","year":"2015","journal-title":"Anesthesiology"},{"issue":"1","key":"pmed.1005024.ref006","doi-asserted-by":"crossref","first-page":"79","DOI":"10.1097\/ALN.0000000000000686","article-title":"Intraoperative mean arterial pressure variability and 30-day mortality in patients having noncardiac surgery","volume":"123","author":"EJ Mascha","year":"2015","journal-title":"Anesthesiology"},{"issue":"2","key":"pmed.1005024.ref007","doi-asserted-by":"crossref","first-page":"307","DOI":"10.1097\/ALN.0000000000000756","article-title":"Association between intraoperative hypotension and hypertension and 30-day postoperative mortality in noncardiac surgery","volume":"123","author":"TG Monk","year":"2015","journal-title":"Anesthesiology"},{"key":"pmed.1005024.ref008","doi-asserted-by":"crossref","first-page":"110516","DOI":"10.1016\/j.jclinane.2021.110516","article-title":"Association of perioperative hypotension with subsequent greater healthcare resource utilization","volume":"75","author":"WH Stapelfeldt","year":"2021","journal-title":"J Clin Anesth"},{"issue":"4","key":"pmed.1005024.ref009","doi-asserted-by":"crossref","first-page":"492","DOI":"10.1007\/s00540-011-1152-1","article-title":"Impact of intraoperative hypotension on hospital stay in major abdominal surgery","volume":"25","author":"V Tassoudis","year":"2011","journal-title":"J Anesth"},{"issue":"6","key":"pmed.1005024.ref010","doi-asserted-by":"crossref","DOI":"10.1016\/j.jval.2023.03.206","article-title":"CO133 clinic and economic burden associated with intraoperative hypotension among high-risk surgical patients in taiwan: a retrospective cohort study","volume":"26","author":"CH Chung","year":"2023","journal-title":"Value in Health"},{"issue":"1","key":"pmed.1005024.ref011","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1186\/s12893-020-01015-z","article-title":"Intraoperative hypotension is not associated with adverse short-term postoperative outcomes after esophagectomy in esophageal cancer patients","volume":"21","author":"ET Yeheyis","year":"2021","journal-title":"BMC Surg"},{"issue":"1","key":"pmed.1005024.ref012","doi-asserted-by":"crossref","first-page":"103","DOI":"10.1186\/s12871-025-02976-5","article-title":"Intraoperative hypotension and postoperative risks in non-cardiac surgery: a meta-analysis","volume":"25","author":"G Qin","year":"2025","journal-title":"BMC Anesthesiology"},{"issue":"1","key":"pmed.1005024.ref013","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1186\/s44158-025-00225-2","article-title":"Single center experience with Hypotension Prediction Index (HPI) during cytoreductive surgery with Hyperthermic Intraperitoneal Chemotherapy (HIPEC)","volume":"5","author":"C L\u2019Acqua","year":"2025","journal-title":"J Anesth Analg Crit Care"},{"issue":"4","key":"pmed.1005024.ref014","doi-asserted-by":"crossref","first-page":"663","DOI":"10.1097\/ALN.0000000000002300","article-title":"Machine-learning algorithm to predict hypotension based on high-fidelity arterial pressure waveform analysis","volume":"129","author":"F Hatib","year":"2018","journal-title":"Anesthesiology"},{"issue":"9","key":"pmed.1005024.ref015","article-title":"Intraoperative hypotension prediction model based on systematic feature engineering and machine learning.","volume":"22","author":"S Lee","year":"2022","journal-title":"Sensors (Basel)"},{"key":"pmed.1005024.ref016","doi-asserted-by":"crossref","DOI":"10.3389\/fanes.2023.1138175","article-title":"Hypotension prediction index: From reactive to predictive hemodynamic management, the key to maintaining hemodynamic stability","volume":"2","author":"J Ripoll\u00e9s-Melchor","year":"2023","journal-title":"Front Anesthesiol"},{"issue":"1","key":"pmed.1005024.ref017","doi-asserted-by":"crossref","first-page":"725","DOI":"10.1186\/s12967-024-05481-4","article-title":"Predictive ability of hypotension prediction index and machine learning methods in intraoperative hypotension: a systematic review and meta-analysis","volume":"22","author":"I Mohammadi","year":"2024","journal-title":"J Transl Med"},{"issue":"1","key":"pmed.1005024.ref018","doi-asserted-by":"crossref","first-page":"661","DOI":"10.1186\/s13054-020-03379-3","article-title":"Prediction of hypotension events with physiologic vital sign signatures in the intensive care unit","volume":"24","author":"JH Yoon","year":"2020","journal-title":"Crit Care"},{"issue":"9","key":"pmed.1005024.ref019","doi-asserted-by":"crossref","first-page":"3108","DOI":"10.3390\/s22093108","article-title":"Intraoperative hypotension prediction model based on systematic feature engineering and machine learning","volume":"22","author":"S Lee","year":"2022","journal-title":"Sensors"},{"key":"pmed.1005024.ref020","doi-asserted-by":"crossref","unstructured":"Thundiyil S, Picone J, McKenzie S. Transformer architectures in time series analysis: a review. 2023;20(1):1\u201310.","DOI":"10.1109\/SPMB59478.2023.10372632"},{"key":"pmed.1005024.ref021","doi-asserted-by":"crossref","unstructured":"Wen Q, Zhou T, Zhang C, Chen W, Ma Z, Yan J. Transformers in time series: a survey. 2022. doi: arXiv:2202.07125","DOI":"10.24963\/ijcai.2023\/759"},{"key":"pmed.1005024.ref022","unstructured":"Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, et al. Attention is all you need. 2017. doi: arXiv:1706.03762"},{"key":"pmed.1005024.ref023","doi-asserted-by":"crossref","unstructured":"Rae JW, Razavi A. Do transformers need deep long-range memory. 2020. Available from: https:\/\/ui.adsabs.harvard.edu\/abs\/2020arXiv200703356R","DOI":"10.18653\/v1\/2020.acl-main.672"},{"issue":"4","key":"pmed.1005024.ref024","doi-asserted-by":"crossref","first-page":"808","DOI":"10.1016\/j.bja.2020.12.035","article-title":"Deep learning models for the prediction of intraoperative hypotension","volume":"126","author":"S Lee","year":"2021","journal-title":"Br J Anaesth"},{"key":"pmed.1005024.ref025","doi-asserted-by":"crossref","DOI":"10.1136\/bmj-2023-078378","article-title":"TRIPOD AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","author":"GS Collins","year":"2024","journal-title":"BMJ"},{"issue":"1","key":"pmed.1005024.ref026","doi-asserted-by":"crossref","first-page":"205","DOI":"10.1186\/cc11455","article-title":"Contrast-induced acute kidney injury and renal support for acute kidney injury: a KDIGO summary (Part 2)","volume":"17","author":"N Lameire","year":"2013","journal-title":"Crit Care"},{"issue":"4","key":"pmed.1005024.ref027","doi-asserted-by":"crossref","first-page":"241","DOI":"10.1038\/nrneph.2017.2","article-title":"Acute kidney disease and renal recovery: consensus report of the Acute Disease Quality Initiative (ADQI) 16 Workgroup","volume":"13","author":"LS Chawla","year":"2017","journal-title":"Nat Rev Nephrol"},{"issue":"3","key":"pmed.1005024.ref028","doi-asserted-by":"crossref","first-page":"658","DOI":"10.1097\/ALN.0b013e3182472320","article-title":"Intraoperative hypotension and perioperative ischemic stroke after general surgery: a nested case-control study","volume":"116","author":"JB Bijker","year":"2012","journal-title":"Anesthesiology"},{"key":"pmed.1005024.ref029","doi-asserted-by":"crossref","first-page":"102797","DOI":"10.1016\/j.eclinm.2024.102797","article-title":"Development and external validation of temporal fusion transformer models for continuous intraoperative blood pressure forecasting","volume":"75","author":"L Kapral","year":"2024","journal-title":"EClinicalMedicine"},{"issue":"1","key":"pmed.1005024.ref030","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1186\/s13040-025-00464-7","article-title":"Short- and long-term weekly patient-reported outcomes prediction undergoing radiotherapy: single-patient time series model vs. transformer-based multi-patient time series model","volume":"18","author":"Y Yan","year":"2025","journal-title":"BioData Min"},{"key":"pmed.1005024.ref031","doi-asserted-by":"crossref","first-page":"105993","DOI":"10.1109\/ACCESS.2022.3211334","article-title":"A multi-headed transformer approach for predicting the patient\u2019s clinical time-series variables from charted vital signs","volume":"10","author":"G Harerimana","year":"2022","journal-title":"IEEE Access"},{"key":"pmed.1005024.ref032","doi-asserted-by":"crossref","first-page":"108060","DOI":"10.1016\/j.cmpb.2024.108060","article-title":"A transformer-based diffusion probabilistic model for heart rate and blood pressure forecasting in Intensive Care Unit","volume":"246","author":"P Chang","year":"2024","journal-title":"Comput Methods Programs Biomed"},{"issue":"1","key":"pmed.1005024.ref033","doi-asserted-by":"crossref","first-page":"279","DOI":"10.1038\/s41597-022-01411-5","article-title":"VitalDB, a high-fidelity multi-parameter vital signs database in surgical patients","volume":"9","author":"H-C Lee","year":"2022","journal-title":"Sci Data"},{"issue":"11","key":"pmed.1005024.ref034","doi-asserted-by":"crossref","first-page":"877","DOI":"10.4103\/ija.IJA_624_19","article-title":"Intraoperative hypotension and its prediction","volume":"63","author":"JJ Vos","year":"2019","journal-title":"Indian J Anaesth"},{"issue":"1","key":"pmed.1005024.ref035","doi-asserted-by":"crossref","first-page":"96","DOI":"10.1186\/s12911-025-02930-y","article-title":"Machine learning-based prediction of post-induction hypotension: identifying risk factors and enhancing anesthesia management","volume":"25","author":"M Chen","year":"2025","journal-title":"BMC Med Inform Decis Mak"},{"issue":"5","key":"pmed.1005024.ref036","doi-asserted-by":"crossref","first-page":"785","DOI":"10.1016\/j.bja.2022.01.032","article-title":"Intraoperative venous congestion rather than hypotension is associated with acute adverse kidney events after cardiac surgery: a retrospective cohort study","volume":"128","author":"L Chen","year":"2022","journal-title":"Br J Anaesth"},{"issue":"4","key":"pmed.1005024.ref037","doi-asserted-by":"crossref","first-page":"639","DOI":"10.1097\/ALN.0000000000005355","article-title":"Hemodynamic management guided by the hypotension prediction index in abdominal surgery: a multicenter randomized clinical trial","volume":"142","author":"J Ripoll\u00e9s-Melchor","year":"2025","journal-title":"Anesthesiology"},{"issue":"9","key":"pmed.1005024.ref038","first-page":"11121","article-title":"Are transformers effective for time series forecasting?","volume":"37","author":"A Zeng","year":"2023","journal-title":"Proc AAAI Conf Artif Intellig"},{"issue":"1","key":"pmed.1005024.ref039","doi-asserted-by":"crossref","first-page":"230","DOI":"10.1186\/s12916-019-1466-7","article-title":"Calibration: the Achilles heel of predictive analytics","volume":"17","author":"Topic Group \u2018Evaluating diagnostic tests and prediction models\u2019 of the STRATOS initiative","year":"2019","journal-title":"BMC Med"},{"issue":"9","key":"pmed.1005024.ref040","doi-asserted-by":"crossref","first-page":"1337","DOI":"10.1038\/s41591-019-0548-6","article-title":"Do no harm: a roadmap for responsible machine learning for health care","volume":"25","author":"J Wiens","year":"2019","journal-title":"Nat Med"}],"updated-by":[{"DOI":"10.1371\/journal.pmed.1005024","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T00:00:00Z","timestamp":1775088000000}}],"container-title":["PLOS Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pmed.1005024","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,2]],"date-time":"2026-04-02T17:51:14Z","timestamp":1775152274000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pmed.1005024"}},"subtitle":[],"editor":[{"given":"Dhiya","family":"Al-Jumeily OBE","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,3,25]]},"references-count":40,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3,25]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pmed.1005024","relation":{},"ISSN":["1549-1676"],"issn-type":[{"value":"1549-1676","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,25]]}}}