{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T05:07:26Z","timestamp":1789103246377,"version":"build-2803163510"},"reference-count":22,"publisher":"MDPI AG","issue":"18","license":[{"start":{"date-parts":[[2026,9,8]],"date-time":"2026-09-08T00:00:00Z","timestamp":1788825600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J. Clin. Med.","JCM"],"abstract":"<jats:p>Background: PADS combines two neural networks predicting ICU mortality and discharge within 48 h, placing critically ill patients into one of four clinically meaningful states. Developed on MIMIC-IV alone, it left open whether it generalizes to other ICUs, whether its models transfer across hospitals, and whether its predictions can be explained at the bedside. Methods: We evaluated PADS on four ICU databases from different hospitals and countries (MIMIC-IV, AmsterdamUMCdb, eICU-CRD, and HiRID), using the same routinely collected variables. Mortality is scored on the final 48-h window (terminal-window, not early-warning, discrimination). For each external database, we compared the MIMIC model used as-is, retrained from scratch, and retrained from the MIMIC weights, and added an explainability layer. Results: For mortality, reusing and retraining the MIMIC model gave the highest discrimination on every database (AUROC 0.955\u20130.986; terminal-window (near-outcome) discrimination) and stabilized training; used as-is, it ranged from chance (Amsterdam) to good (eICU, HiRID). For discharge, training fresh on local data matched or beat reusing MIMIC on every external database, consistent with discharge timing depending on local organization rather than physiology. The explainability layer produced clinically coherent, cross-checked explanations. Conclusions: Transportability was task-dependent: mortality transferred between hospitals, discharge did not. PADS demonstrated promising external transportability across heterogeneous ICU databases, particularly after local adaptation. Reusing and adapting the MIMIC-IV mortality model across hospitals improves accuracy. This approach also stabilizes training, providing a basis for potential federated deployment, whereas discharge is better trained locally. The mortality results reported here are terminal-window discrimination and do not support use of the framework as an early-warning model. A transparent explainability layer provides an interpretable representation of model predictions, addressing a key barrier to clinical adoption.<\/jats:p>","DOI":"10.3390\/jcm15186957","type":"journal-article","created":{"date-parts":[[2026,9,8]],"date-time":"2026-09-08T13:46:33Z","timestamp":1788875193000},"page":"6957","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Generalization, Cross-ICU Transfer, and Explainability of a Mortality and Time-to-Discharge Framework for the Intensive Care Unit"],"prefix":"10.3390","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0230-0464","authenticated-orcid":false,"given":"\u00c0lex","family":"Pardo","sequence":"first","affiliation":[{"name":"Department of Medicine and Surgery, Rovira Virgili University, 43201 Tarragona, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0573-7621","authenticated-orcid":false,"given":"Josep","family":"G\u00f3mez","sequence":"additional","affiliation":[{"name":"Department of Medicine and Surgery, Rovira Virgili University, 43201 Tarragona, Spain"},{"name":"Critical Care Department, Joan XXIII University Hospital, 43005 Tarragona, Spain"},{"name":"Pere Virgili Health Research Institute, Joan XXIII University Hospital, Rovira Virgili University, 43005 Tarragona, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-1015-9780","authenticated-orcid":false,"given":"Julen","family":"Berrueta","sequence":"additional","affiliation":[{"name":"Department of Medicine and Surgery, Rovira Virgili University, 43201 Tarragona, Spain"},{"name":"Critical Care Department, Joan XXIII University Hospital, 43005 Tarragona, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-0900-5742","authenticated-orcid":false,"given":"Alejandro","family":"Garc\u00eda-Mart\u00ednez","sequence":"additional","affiliation":[{"name":"Critical Care Department, Joan XXIII University Hospital, 43005 Tarragona, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-2304-1022","authenticated-orcid":false,"given":"Pau","family":"Orts","sequence":"additional","affiliation":[{"name":"Department of Medicine and Surgery, Rovira Virgili University, 43201 Tarragona, Spain"},{"name":"Critical Care Department, Joan XXIII University Hospital, 43005 Tarragona, Spain"},{"name":"Pere Virgili Health Research Institute, Joan XXIII University Hospital, Rovira Virgili University, 43005 Tarragona, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8554-7598","authenticated-orcid":false,"given":"Sara","family":"Manrique","sequence":"additional","affiliation":[{"name":"Department of Medicine and Surgery, Rovira Virgili University, 43201 Tarragona, Spain"},{"name":"Critical Care Department, Joan XXIII University Hospital, 43005 Tarragona, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8828-5984","authenticated-orcid":false,"given":"Alejandro","family":"Rodr\u00edguez","sequence":"additional","affiliation":[{"name":"Department of Medicine and Surgery, Rovira Virgili University, 43201 Tarragona, Spain"},{"name":"Critical Care Department, Joan XXIII University Hospital, 43005 Tarragona, Spain"},{"name":"Pere Virgili Health Research Institute, Joan XXIII University Hospital, Rovira Virgili University, 43005 Tarragona, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7652-8379","authenticated-orcid":false,"given":"Mar\u00eda","family":"Bod\u00ed","sequence":"additional","affiliation":[{"name":"Department of Medicine and Surgery, Rovira Virgili University, 43201 Tarragona, Spain"},{"name":"Critical Care Department, Joan XXIII University Hospital, 43005 Tarragona, Spain"},{"name":"Pere Virgili Health Research Institute, Joan XXIII University Hospital, Rovira Virgili University, 43005 Tarragona, Spain"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,9,8]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Pardo, \u00c0., G\u00f3mez, J., Berrueta, J., Garc\u00eda, A., Manrique, S., Rodr\u00edguez, A., and Bod\u00ed, M. (2025). Combining Predictive Models of Mortality and Time-to-Discharge for Improved Outcome Assessment in Intensive Care Units. J. Clin. Med., 14.","DOI":"10.3390\/jcm14134515"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"e489","DOI":"10.1016\/S2589-7500(20)30186-2","article-title":"The myth of generalisability in clinical research and machine learning in health care","volume":"2","author":"Futoma","year":"2020","journal-title":"Lancet Digit. Health"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"1065","DOI":"10.1001\/jamainternmed.2021.2626","article-title":"External Validation of a Widely Implemented Proprietary Sepsis Prediction Model in Hospitalized Patients","volume":"181","author":"Wong","year":"2021","journal-title":"JAMA Intern. Med."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"49","DOI":"10.1093\/ckj\/sfaa188","article-title":"External validation of prognostic models: What, why, how, when and where?","volume":"14","author":"Ramspek","year":"2021","journal-title":"Clin. Kidney J."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"76","DOI":"10.1186\/s13054-024-04866-7","article-title":"Real-time machine learning model to predict short-term mortality in critically ill patients: Development and international validation","volume":"28","author":"Lim","year":"2024","journal-title":"Crit. Care"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41597-022-01899-x","article-title":"MIMIC-IV, a freely accessible electronic health record dataset","volume":"10","author":"Johnson","year":"2023","journal-title":"Sci. Data"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"e563","DOI":"10.1097\/CCM.0000000000004916","article-title":"Sharing ICU Patient Data Responsibly Under the Society of Critical Care Medicine\/European Society of Intensive Care Medicine Joint Data Science Collaboration: The Amsterdam University Medical Centers Database (AmsterdamUMCdb) Example","volume":"49","author":"Thoral","year":"2021","journal-title":"Crit. Care Med."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"180178","DOI":"10.1038\/sdata.2018.178","article-title":"The eICU Collaborative Research Database, a freely available multi-center database for critical care research","volume":"5","author":"Pollard","year":"2018","journal-title":"Sci. Data"},{"key":"ref_9","unstructured":"Faltys, M., Zimmermann, M., Lyu, X., H\u00fcser, M., Hyland, S., R\u00e4tsch, G., and Merz, T. (2021). HiRID, a high time-resolution ICU dataset (version 1.1.1). PhysioNet."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"R161","DOI":"10.1186\/cc7160","article-title":"Evaluation of SOFA-based models for predicting mortality in the ICU: A systematic review","volume":"12","author":"Minne","year":"2008","journal-title":"Crit. Care"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"119","DOI":"10.1038\/s41746-020-00323-1","article-title":"The future of digital health with federated learning","volume":"3","author":"Rieke","year":"2020","journal-title":"NPJ Digit. Med."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"0321","DOI":"10.34133\/hds.0321","article-title":"Bridging data gaps in healthcare: A scoping review of transfer learning in structured data analysis","volume":"5","author":"Li","year":"2025","journal-title":"Health Data Sci."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"985","DOI":"10.1097\/CCM.0000000000005894","article-title":"Bringing the Promise of Artificial Intelligence to Critical Care: What the Experience With Sepsis Analytics Can Teach Us","volume":"51","author":"Wardi","year":"2023","journal-title":"Crit. Care Med."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"163","DOI":"10.1377\/hlthaff.2024.01003","article-title":"Artificial Intelligence In Health And Health Care: Priorities For Action","volume":"44","author":"Matheny","year":"2025","journal-title":"Health Aff."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"364","DOI":"10.1038\/s41591-020-0789-4","article-title":"Early prediction of circulatory failure in the intensive care unit using machine learning","volume":"26","author":"Hyland","year":"2020","journal-title":"Nat. Med."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"E215","DOI":"10.1161\/01.CIR.101.23.e215","article-title":"PhysioBank, PhysioToolkit, and PhysioNet","volume":"101","author":"Goldberger","year":"2000","journal-title":"Circulation"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","article-title":"A Survey on Transfer Learning","volume":"22","author":"Pan","year":"2010","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_18","first-page":"3320","article-title":"How transferable are features in deep neural networks?","volume":"27","author":"Yosinski","year":"2014","journal-title":"Adv. Neural Inf. Process. Syst. (NeurIPS)"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K. (2019). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. Proceedings of NAACL-HLT 2019, Association for Computational Linguistics.","DOI":"10.18653\/v1\/N19-1423"},{"key":"ref_20","unstructured":"Sundararajan, M., Taly, A., and Yan, Q. (2026, September 02). Axiomatic Attribution for Deep Networks (Integrated Gradients). ICML 2017. Available online: https:\/\/proceedings.mlr.press\/v70\/sundararajan17a.html."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"818","DOI":"10.1007\/978-3-319-10590-1_53","article-title":"Visualizing and Understanding Convolutional Networks","volume":"Volume 8689","author":"Zeiler","year":"2014","journal-title":"Computer Vision\u2014ECCV 2014"},{"key":"ref_22","unstructured":"Lundberg, S.M., and Lee, S.-I. (2026, September 02). A Unified Approach to Interpreting Model Predictions (SHAP). NeurIPS 2017. Available online: https:\/\/proceedings.neurips.cc\/paper\/2017\/hash\/8a20a8621978632d76c43dfd28b67767-Abstract.html."}],"container-title":["Journal of Clinical Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2077-0383\/15\/18\/6957\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T04:34:46Z","timestamp":1789101286000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2077-0383\/15\/18\/6957"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,8]]},"references-count":22,"journal-issue":{"issue":"18","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["jcm15186957"],"URL":"https:\/\/doi.org\/10.3390\/jcm15186957","relation":{},"ISSN":["2077-0383"],"issn-type":[{"value":"2077-0383","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,8]]}}}