{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T15:13:24Z","timestamp":1785510804617,"version":"3.56.0"},"reference-count":28,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T00:00:00Z","timestamp":1785456000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Digit. Health"],"abstract":"<jats:p>ICU delirium is a common and serious complication in critically ill patients, yet existing prediction models rely predominantly on static data and lack interpretable decision logic, limiting their clinical utility. This retrospective cohort study enrolled 464 ICU patients from the First Affiliated Hospital of Zunyi Medical University as the internal cohort and 70 patients from the Second Affiliated Hospital as an external test set. A variable system comprising nine baseline characteristics and 24 dynamic variables was established through systematic evidence synthesis and a clinical pilot investigation. Dynamic variables were organized into 24-hour rolling observation windows (T1\u2013T7), and a two-layer bidirectional long short-term memory (LSTM) network incorporating a masking mechanism and class-weight correction was developed to predict delirium occurrence in each window based on all preceding sequential inputs. The model achieved an area under the receiver operating characteristic curve (AUCROC) of 0.835 (95% CI: 0.724\u20130.946), sensitivity of 0.696, and specificity of 0.894 on the internal test set, with satisfactory calibration and clinical net benefit on this set across a broad probability threshold range. External validation yielded an AUCROC of 0.830 (95% CI: 0.716\u20130.944), with calibration and decision curve analysis in the external cohort further supporting reliable probability estimates and clinical net benefit, indicating stable cross-institutional generalizability. SHapley Additive exPlanations (SHAP) analysis revealed a clinically coherent, phase-dependent transition in dominant predictors: invasive mechanical ventilation predominated during T1\u2013T3 (peak mean |SHAP|\u2009=\u20090.286 at T3), arterial pH and Sequential Organ Failure Assessment (SOFA) score emerged as principal drivers during T4\u2013T6, and SOFA score retained the highest attribution at T7, collectively providing an evidence-based rationale for temporally differentiated monitoring and intervention strategies.<\/jats:p>","DOI":"10.3389\/fdgth.2026.1856311","type":"journal-article","created":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T14:55:54Z","timestamp":1785509754000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A long short-term memory network with SHAP interpretability for dynamic prediction of ICU delirium: development and external validation"],"prefix":"10.3389","volume":"8","author":[{"given":"Lanqiong","family":"Lei","sequence":"first","affiliation":[{"name":"Department of Intensive Care Unit, The Affiliated Hospital of Zunyi Medical University","place":["Zunyi, Guizhou, China"]},{"name":"Nursing Department, The Affiliated Hospital of Zunyi Medical University","place":["Zunyi, Guizhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongya","family":"Xia","sequence":"additional","affiliation":[{"name":"Nursing Department, The Affiliated Hospital of Zunyi Medical University","place":["Zunyi, Guizhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ran","family":"Zhang","sequence":"additional","affiliation":[{"name":"Zunyi Medical University","place":["Zunyi, Guizhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yu","family":"Huang","sequence":"additional","affiliation":[{"name":"Zunyi Medical University","place":["Zunyi, Guizhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fang","family":"He","sequence":"additional","affiliation":[{"name":"Department of Neurology, The Second Affiliated Hospital of Zunyi Medical University","place":["Zunyi, Guizhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shuwen","family":"Huang","sequence":"additional","affiliation":[{"name":"Zunyi Medical University","place":["Zunyi, Guizhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xusihong","family":"Cai","sequence":"additional","affiliation":[{"name":"Zunyi Medical University","place":["Zunyi, Guizhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Juan","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Joint Surgery, Tongren People\u2019s Hospital","place":["Tongren, Guizhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jing","family":"Zhou","sequence":"additional","affiliation":[{"name":"Nursing Department, The Affiliated Hospital of Zunyi Medical University","place":["Zunyi, Guizhou, China"]},{"name":"School of Nursing, Zunyi Medical University","place":["Zunyi, Guizhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,31]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"2093","DOI":"10.1007\/s00134-025-08167-7","article-title":"Prevalence and incidence of ICU delirium and pain: a systematic review and meta-analysis","volume":"51","author":"Leong","year":"2025","journal-title":"Intensive Care Med"},{"key":"B2","doi-asserted-by":"publisher","first-page":"e2235339","DOI":"10.1001\/jamanetworkopen.2022.35339","article-title":"Association of incident delirium with short-term mortality in adults with critical illness receiving mechanical ventilation","volume":"5","author":"Li","year":"2022","journal-title":"JAMA Netw Open"},{"key":"B3","doi-asserted-by":"publisher","first-page":"387","DOI":"10.34197\/ats-scholar.2022-0115OT","article-title":"Delirium and long-term cognitive impairment after critical illness","volume":"4","author":"Mart","year":"2023","journal-title":"ATS Sch"},{"key":"B4","doi-asserted-by":"publisher","first-page":"103782","DOI":"10.1016\/j.ijnurstu.2020.103782","article-title":"Diagnostic accuracy of the CAM-ICU and ICDSC in detecting intensive care unit delirium: a bivariate meta-analysis","volume":"113","author":"Chen","year":"2021","journal-title":"Int J Nurs Stud"},{"key":"B5","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1186\/s12912-025-02704-x","article-title":"Barriers to nurse-led delirium management in intensive care units: an integrative systematic review using COM-B model","volume":"24","author":"Almoliky","year":"2025","journal-title":"BMC Nurs"},{"key":"B6","doi-asserted-by":"publisher","first-page":"1655","DOI":"10.1016\/j.jamda.2022.06.020","article-title":"Machine learning-based prediction models for delirium: a systematic review and meta-analysis","volume":"23","author":"Xie","year":"2022","journal-title":"J Am Med Dir Assoc"},{"key":"B7","doi-asserted-by":"publisher","first-page":"e0296","DOI":"10.1097\/CCE.0000000000000296","article-title":"ICU delirium-prediction models: a systematic review","volume":"2","author":"Ruppert","year":"2020","journal-title":"Crit Care Explor"},{"key":"B8","doi-asserted-by":"publisher","first-page":"634","DOI":"10.3390\/s23020634","article-title":"Survey of explainable AI techniques in healthcare","volume":"23","author":"Chaddad","year":"2023","journal-title":"Sensors"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1235","DOI":"10.1162\/neco_a_01199","article-title":"A review of recurrent neural networks: LSTM cells and network architectures","volume":"31","author":"Yu","year":"2019","journal-title":"Neural Comput"},{"key":"B10","doi-asserted-by":"publisher","first-page":"933037","DOI":"10.3389\/fmed.2022.933037","article-title":"Explainable time-series deep learning models for the prediction of mortality, prolonged length of stay and 30-day readmission in intensive care patients","volume":"9","author":"Deng","year":"2022","journal-title":"Front Med"},{"key":"B11","doi-asserted-by":"publisher","first-page":"ooaf065","DOI":"10.1093\/jamiaopen\/ooaf065","article-title":"Artificial intelligence models for predicting acute kidney injury in the intensive care unit: a systematic review of modeling methods, data utilization, and clinical applicability","volume":"8","author":"Shi","year":"2025","journal-title":"JAMIA Open"},{"key":"B12","doi-asserted-by":"publisher","first-page":"239","DOI":"10.3390\/diagnostics14030239","article-title":"An improved long short-term memory algorithm for cardiovascular disease prediction","volume":"14","author":"Revathi","year":"2024","journal-title":"Diagnostics"},{"key":"B13","doi-asserted-by":"publisher","first-page":"e70056","DOI":"10.1111\/cts.70056","article-title":"Practical guide to SHAP analysis: explaining supervised machine learning model predictions in drug development","volume":"17","author":"Ponce-Bobadilla","year":"2024","journal-title":"Clin Transl Sci"},{"key":"B14","doi-asserted-by":"publisher","first-page":"E839","DOI":"10.1503\/cmaj.090449","article-title":"AGREE II: advancing guideline development, reporting and evaluation in health care","volume":"182","author":"Brouwers","year":"2010","journal-title":"CMAJ"},{"key":"B15","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1186\/s13643-024-02622-0","article-title":"Methodological and reporting quality assessment of systematic reviews and meta-analyses in the association between sleep duration and hypertension","volume":"13","author":"Yang","year":"2024","journal-title":"Syst Rev"},{"key":"B16","doi-asserted-by":"publisher","first-page":"603","DOI":"10.1007\/s10654-010-9491-z","article-title":"Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses","volume":"25","author":"Stang","year":"2010","journal-title":"Eur J Epidemiol"},{"key":"B17","doi-asserted-by":"publisher","first-page":"51","DOI":"10.7326\/M18-1376","article-title":"PROBAST: a tool to assess the risk of bias and applicability of prediction model studies","volume":"170","author":"Wolff","year":"2019","journal-title":"Ann Intern Med"},{"key":"B18","doi-asserted-by":"publisher","first-page":"299","DOI":"10.1097\/ALN.0000000000004478","article-title":"Predicting intensive care delirium with machine learning: model development and external validation","volume":"138","author":"Gong","year":"2023","journal-title":"Anesthesiology"},{"key":"B19","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/bhi58575.2023.10313445","article-title":"Dynamic delirium prediction in the intensive care unit using machine learning on electronic health records","volume":"2023","author":"Contreras","year":"2023","journal-title":"IEEE EMBS Int Conf Biomed Health Inform"},{"key":"B20","doi-asserted-by":"publisher","first-page":"e23401","DOI":"10.2196\/23401","article-title":"A machine learning-based algorithm for the prediction of intensive care unit delirium (PRIDE): retrospective study","volume":"9","author":"Hur","year":"2021","journal-title":"JMIR Med Inform"},{"key":"B21","doi-asserted-by":"publisher","first-page":"2570","DOI":"10.1109\/EMBC.2017.8037382","article-title":"Combination of static and temporal data analysis to predict mortality and readmission in the intensive care","volume":"2017","author":"Venugopalan","year":"2017","journal-title":"Annu Int Conf IEEE Eng Med Biol Soc"},{"key":"B22","doi-asserted-by":"publisher","first-page":"34","DOI":"10.1016\/j.jcrc.2020.10.006","article-title":"On predictions in critical care: the individual prognostication fallacy in elderly patients","volume":"61","author":"Beil","year":"2021","journal-title":"J Crit Care"},{"key":"B23","doi-asserted-by":"publisher","first-page":"e825","DOI":"10.1097\/CCM.0000000000003299","article-title":"Clinical practice guidelines for the prevention and management of pain, agitation\/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU","volume":"46","author":"Devlin","year":"2018","journal-title":"Crit Care Med"},{"key":"B24","doi-asserted-by":"publisher","first-page":"1446523","DOI":"10.3389\/fnagi.2024.1446523","article-title":"Mechanisms underlying delirium in patients with critical illness","volume":"16","author":"Fan","year":"2024","journal-title":"Front Aging Neurosci"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1179243","DOI":"10.3389\/fneur.2023.1179243","article-title":"Meta-analysis of serological biomarkers at hospital admission for the likelihood of developing delirium during hospitalization","volume":"14","author":"Bassi","year":"2023","journal-title":"Front Neurol"},{"key":"B26","doi-asserted-by":"publisher","first-page":"30265","DOI":"10.1038\/s41598-024-81559-9","article-title":"High level of systemic immune inflammation index elevates delirium risk among patients in intensive care unit","volume":"14","author":"Xu","year":"2024","journal-title":"Sci Rep"},{"key":"B27","doi-asserted-by":"publisher","first-page":"542","DOI":"10.1016\/j.csbj.2024.08.005","article-title":"Unveiling the black box: a systematic review of explainable artificial intelligence in medical image analysis","volume":"24","author":"Muhammad","year":"2024","journal-title":"Comput Struct Biotechnol J"},{"key":"B28","doi-asserted-by":"publisher","first-page":"29","DOI":"10.1016\/j.inffus.2021.07.016","article-title":"Unbox the black-box for the medical explainable AI via multi-modal and multi-centre data fusion: a mini-review, two showcases and beyond","volume":"77","author":"Yang","year":"2022","journal-title":"Inf Fusion"}],"container-title":["Frontiers in Digital Health"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1856311\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T14:55:55Z","timestamp":1785509755000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fdgth.2026.1856311\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,31]]},"references-count":28,"alternative-id":["10.3389\/fdgth.2026.1856311"],"URL":"https:\/\/doi.org\/10.3389\/fdgth.2026.1856311","relation":{},"ISSN":["2673-253X"],"issn-type":[{"value":"2673-253X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,31]]},"article-number":"1856311"}}