{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,13]],"date-time":"2026-11-13T09:35:57Z","timestamp":1794562557305,"version":"4.9.1"},"reference-count":87,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T00:00:00Z","timestamp":1781136000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T00:00:00Z","timestamp":1790208000000},"content-version":"vor","delay-in-days":105,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100018693","name":"HORIZON EUROPE Framework Programme","doi-asserted-by":"publisher","award":["Grant Agreement No 101136244 (TARGET)"],"award-info":[{"award-number":["Grant Agreement No 101136244 (TARGET)"]}],"id":[{"id":"10.13039\/100018693","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100018693","name":"HORIZON EUROPE Framework Programme","doi-asserted-by":"publisher","award":["Grant Agreement No 101136244 (TARGET)"],"award-info":[{"award-number":["Grant Agreement No 101136244 (TARGET)"]}],"id":[{"id":"10.13039\/100018693","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100018693","name":"HORIZON EUROPE Framework Programme","doi-asserted-by":"publisher","award":["Grant Agreement No 101136244 (TARGET)"],"award-info":[{"award-number":["Grant Agreement No 101136244 (TARGET)"]}],"id":[{"id":"10.13039\/100018693","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100018693","name":"HORIZON EUROPE Framework Programme","doi-asserted-by":"publisher","award":["Grant Agreement No 101136244 (TARGET)"],"award-info":[{"award-number":["Grant Agreement No 101136244 (TARGET)"]}],"id":[{"id":"10.13039\/100018693","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100018693","name":"HORIZON EUROPE Framework Programme","doi-asserted-by":"publisher","award":["Grant Agreement No 101136244 (TARGET)"],"award-info":[{"award-number":["Grant Agreement No 101136244 (TARGET)"]}],"id":[{"id":"10.13039\/100018693","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100018693","name":"HORIZON EUROPE Framework Programme","doi-asserted-by":"publisher","award":["Grant Agreement No 101136244 (TARGET)"],"award-info":[{"award-number":["Grant Agreement No 101136244 (TARGET)"]}],"id":[{"id":"10.13039\/100018693","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Digit. Med."],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Critically ill patients frequently require multiple concurrent interventions with complex interdependencies, yet existing prediction models treat these as independent events. We developed FIRST-ICU (Forecasting Interventions and Risk Stratification in the ICU), a unified deep learning framework integrating a graph neural network encoder, LSTM, and a novel Intervention Interaction Attention Module (IIAM) for joint prediction of seven ICU interventions. FIRST-ICU was developed using MIMIC-IV (\n                    <jats:italic>n<\/jats:italic>\n                    \u2009=\u200923,926) and externally validated on AmsterdamUMCdb (\n                    <jats:italic>n<\/jats:italic>\n                    \u2009=\u200912,603) without retraining. The Temporal Decoder achieved AUC-ROC\u2009&gt;\u20090.98 for all interventions with Brier scores &lt;0.035. The Discrete Decoder outperformed state-of-the-art baselines, achieving the highest Macro Average Precision for six of seven interventions. Gains were largest for vasopressor interventions, with norepinephrine and phenylephrine improving by 28.0% and 32.8%, respectively, over CNN baselines. Ablation analysis showed that IIAM improved AUC-PR, particularly for vasopressor prediction. Cold-start analysis confirmed robust prediction from physiological signals alone (AUC-PR 0.667\u20130.868). External validation maintained AUC-ROC\u2009&gt;\u20090.90 across intervention categories despite substantial differences in prescribing practices. A Generative Topographic Mapping layer identified six clinically distinct phenotypes, enabling interpretable risk stratification. FIRST-ICU advances multi-intervention prediction through joint modelling of treatment co-occurrence patterns, validated physiological learning, cross-continental generalisability, and interpretable risk stratification. It provides a framework for multi-intervention ICU decision support.\n                  <\/jats:p>","DOI":"10.1038\/s41746-026-02890-1","type":"journal-article","created":{"date-parts":[[2026,6,11]],"date-time":"2026-06-11T05:33:00Z","timestamp":1781155980000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["FIRST-ICU: forecasting interventions and risk stratification in the ICU using graph neural network autoencoders"],"prefix":"10.1038","volume":"9","author":[{"given":"Nosa","family":"Aikodon","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ivan","family":"Olier","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Brian W.","family":"Johnston","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ingeborg D.","family":"Welters","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gregory Y. H.","family":"Lip","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sandra","family":"Ortega-Martorell","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,6,11]]},"reference":[{"key":"2890_CR1","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s12871-022-01777-4","volume":"22","author":"T Carval","year":"2022","unstructured":"Carval, T. et al. Outcomes of patients admitted to the ICU for acute stroke: a retrospective cohort. BMC Anesthesiol. 22, 1\u20138 (2022).","journal-title":"BMC Anesthesiol."},{"key":"2890_CR2","doi-asserted-by":"publisher","unstructured":"Ortega-Martorell, S., Olier, I., Ohlsson, M. & Lip, G. Y. H. Advancing personalised care in atrial fibrillation and stroke: The potential impact of AI from prevention to rehabilitation. Trends Cardiovasc. Med. https:\/\/doi.org\/10.1016\/J.TCM.2024.12.003. (2024).","DOI":"10.1016\/J.TCM.2024.12.003"},{"key":"2890_CR3","doi-asserted-by":"publisher","first-page":"2300","DOI":"10.1111\/jocn.14184","volume":"27","author":"C Minton","year":"2018","unstructured":"Minton, C., Batten, L. & Huntington, A. The impact of a prolonged stay in the ICU on patients\u2019 fundamental care needs. J. Clin. Nurs. 27, 2300\u20132310 (2018).","journal-title":"J. Clin. Nurs."},{"key":"2890_CR4","doi-asserted-by":"publisher","first-page":"u211734.w4726","DOI":"10.1136\/bmjquality.u211734.w4726","volume":"5","author":"Z van Willigen","year":"2016","unstructured":"van Willigen, Z., Collings, N., Richardson, D. & Cusack, R. Quality improvement: the delivery of true early mobilisation in an intensive care unit. BMJ Qual. Improv. Rep. 5, u211734.w4726 (2016).","journal-title":"BMJ Qual. Improv. Rep."},{"key":"2890_CR5","doi-asserted-by":"publisher","first-page":"707","DOI":"10.1007\/BF01709751","volume":"22","author":"JL Vincent","year":"1996","unstructured":"Vincent, J. L. et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction\/failure. Intensive Care Med. 22, 707\u2013710 (1996).","journal-title":"Intensive Care Med."},{"key":"2890_CR6","doi-asserted-by":"publisher","first-page":"e893","DOI":"10.1016\/S2589-7500(22)00154-6","volume":"4","author":"CM Sauer","year":"2022","unstructured":"Sauer, C. M. et al. Leveraging electronic health records for data science: common pitfalls and how to avoid them. Lancet Digit. Health 4, e893\u2013e898 (2022).","journal-title":"Lancet Digit. Health"},{"key":"2890_CR7","doi-asserted-by":"publisher","first-page":"1700","DOI":"10.1093\/cvr\/cvab169","volume":"117","author":"I Olier","year":"2021","unstructured":"Olier, I., Ortega-Martorell, S., Pieroni, M. & Lip, G. Y. H. How machine learning is impacting research in atrial fibrillation: implications for risk prediction and future management. Cardiovasc. Res. 117, 1700\u20131717 (2021).","journal-title":"Cardiovasc. Res."},{"key":"2890_CR8","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2018.178","volume":"5","author":"TJ Pollard","year":"2018","unstructured":"Pollard, T. J. et al. The eICU collaborative research database, a freely available multi-center database for critical care research. Sci. Data 5, 180178 (2018).","journal-title":"Sci. Data"},{"key":"2890_CR9","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-022-01899-x","volume":"10","author":"AEW Johnson","year":"2023","unstructured":"Johnson, A. E. W. et al. MIMIC-IV, a freely accessible electronic health record dataset. Sci. Data 10, 1 (2023).","journal-title":"Sci. Data"},{"key":"2890_CR10","doi-asserted-by":"publisher","first-page":"E563","DOI":"10.1097\/CCM.0000000000004916","volume":"49","author":"PJ Thoral","year":"2021","unstructured":"Thoral, P. J. et al. 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. Crit. Care Med. 49, E563\u2013E577 (2021).","journal-title":"Crit. Care Med."},{"key":"2890_CR11","doi-asserted-by":"publisher","first-page":"1230854","DOI":"10.3389\/fmed.2023.1230854","volume":"10","author":"S Ortega-Martorell","year":"2023","unstructured":"Ortega-Martorell, S., Olier, I., Johnston, B. W. & Welters, I. D. Sepsis-induced coagulopathy is associated with new episodes of atrial fibrillation in patients admitted to critical care in sinus rhythm. Front. Med. 10, 1230854 (2023).","journal-title":"Front. Med."},{"key":"2890_CR12","doi-asserted-by":"publisher","first-page":"915224","DOI":"10.3389\/fmed.2022.915224","volume":"9","author":"M Pieroni","year":"2022","unstructured":"Pieroni, M., Olier, I., Ortega-Martorell, S., Johnston, B. W. & Welters, I. D. In-hospital mortality of sepsis differs depending on the origin of infection: an investigation of predisposing factors. Front. Med. 9, 915224 (2022).","journal-title":"Front. Med."},{"key":"2890_CR13","doi-asserted-by":"publisher","first-page":"897709","DOI":"10.3389\/fcvm.2022.897709","volume":"9","author":"S Ortega-Martorell","year":"2022","unstructured":"Ortega-Martorell, S., Pieroni, M., Johnston, B. W., Olier, I. & Welters, I. D. Development of a risk prediction model for new episodes of atrial fibrillation in medical-surgical critically Ill patients using the AmsterdamUMCdb. Front. Cardiovasc. Med. 9, 897709 (2022).","journal-title":"Front. Cardiovasc. Med."},{"key":"2890_CR14","doi-asserted-by":"publisher","first-page":"6","DOI":"10.3390\/a17010006","volume":"17","author":"N Aikodon","year":"2023","unstructured":"Aikodon, N., Ortega-Martorell, S. & Olier, I. Predicting decompensation risk in intensive care unit patients using machine learning. Algorithms 17, 6 (2023).","journal-title":"Algorithms"},{"key":"2890_CR15","doi-asserted-by":"publisher","first-page":"557","DOI":"10.1177\/0885066616678394","volume":"33","author":"BJ Daly","year":"2018","unstructured":"Daly, B. J. et al. Complexity analysis of decision-making in the critically Ill. J. Intensive Care Med. 33, 557\u2013566 (2018).","journal-title":"J. Intensive Care Med."},{"key":"2890_CR16","doi-asserted-by":"publisher","DOI":"10.1186\/s13054-020-2785-y","volume":"24","author":"G Gutierrez","year":"2020","unstructured":"Gutierrez, G. Artificial intelligence in the intensive care unit. Crit. Care 24, 101 (2020).","journal-title":"Crit. Care"},{"key":"2890_CR17","unstructured":"Lipton, Z. C., Kale, D. C., Elkan, C. & Wetzel, R. Learning to diagnose with LSTM recurrent neural networks. In Proc. 4th International Conference on Learning Representations, ICLR 2016 - Conference Track Proceedings https:\/\/arxiv.org\/pdf\/1511.03677 (2015)."},{"key":"2890_CR18","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","volume":"9","author":"S Hochreiter","year":"1997","unstructured":"Hochreiter, S. & Schmidhuber, J. Long short-term memory. Neural Comput. 9, 1735\u20131780 (1997).","journal-title":"Neural Comput."},{"key":"2890_CR19","doi-asserted-by":"publisher","unstructured":"Sainath, T. N., Vinyals, O., Senior, A. & Sak, H. Convolutional, Long Short-Term Memory, fully connected Deep Neural Networks. in Proc. 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 4580\u20134584 (2015). https:\/\/doi.org\/10.1109\/ICASSP.2015.7178838.","DOI":"10.1109\/ICASSP.2015.7178838"},{"key":"2890_CR20","doi-asserted-by":"publisher","first-page":"152","DOI":"10.1016\/j.eswa.2018.12.055","volume":"122","author":"D Balderas","year":"2019","unstructured":"Balderas, D., Ponce, P. & Molina, A. Convolutional long short term memory deep neural networks for image sequence prediction. Expert Syst. Appl. 122, 152\u2013162 (2019).","journal-title":"Expert Syst. Appl."},{"key":"2890_CR21","first-page":"4091","volume":"32","author":"H Song","year":"2018","unstructured":"Song, H., Rajan, D., Thiagarajan, J. J. & Spanias, A. Attend and diagnose: clinical time series analysis using attention models. Proc. AAAI Conf. Artif. Intell. 32, 4091\u20134098 (2018).","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"2890_CR22","doi-asserted-by":"publisher","first-page":"3709","DOI":"10.1109\/JBHI.2024.3379998","volume":"28","author":"Z Xu","year":"2024","unstructured":"Xu, Z. et al. Predicting ICU interventions: a transparent decision support model based on multivariate time series graph convolutional neural network. IEEE J. Biomed. Health Inform. 28, 3709\u20133720 (2024).","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"2890_CR23","unstructured":"Kipf, T. N. & Welling, M. Semi-Supervised Classification with Graph Convolutional Networks. In Proc. 5th International Conference on Learning Representations, ICLR 2017 - Conference Track Proceedings https:\/\/arxiv.org\/pdf\/1609.02907 (2016)."},{"key":"2890_CR24","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-020-0253-3","volume":"3","author":"MP Sendak","year":"2020","unstructured":"Sendak, M. P., Gao, M., Brajer, N. & Balu, S. Presenting machine learning model information to clinical end users with model facts labels. npj Digital Med. 3, 41 (2020).","journal-title":"npj Digital Med."},{"key":"2890_CR25","first-page":"359","volume":"106","author":"S Tonekaboni","year":"2019","unstructured":"Tonekaboni, S., Joshi, S., McCradden, M. D. & Goldenberg, A. What clinicians want: contextualizing explainable machine learning for clinical end use. Proc. Mach. Learn. Res. 106, 359\u2013380 (2019).","journal-title":"Proc. Mach. Learn. Res."},{"key":"2890_CR26","doi-asserted-by":"publisher","DOI":"10.1186\/s13040-022-00309-7","volume":"15","author":"Z Zeng","year":"2022","unstructured":"Zeng, Z., Tang, X., Liu, Y., He, Z. & Gong, X. Interpretable recurrent neural network models for dynamic prediction of the extubation failure risk in patients with invasive mechanical ventilation in the intensive care unit. BioData Min. 15, 21 (2022).","journal-title":"BioData Min."},{"key":"2890_CR27","doi-asserted-by":"publisher","first-page":"73","DOI":"10.1097\/SHK.0000000000001692","volume":"56","author":"GH Kwak","year":"2021","unstructured":"Kwak, G. H., Ling, L. & Hui, P. Predicting the need for vasopressors in the intensive care unit using an attention based deep learning model. Shock 56, 73\u201379 (2021).","journal-title":"Shock"},{"key":"2890_CR28","first-page":"82","volume":"2017","author":"M Ghassemi","year":"2017","unstructured":"Ghassemi, M., Wu, M., Hughes, M. C., Szolovits, P. & Doshi-Velez, F. Predicting intervention onset in the ICU with switching state space models. AMIA Jt. Summits Transl. Sci. Proc. 2017, 82\u201391 (2017).","journal-title":"AMIA Jt. Summits Transl. Sci. Proc."},{"key":"2890_CR29","unstructured":"Suresh, H. et al. Clinical intervention prediction and understanding with deep neural networks. In Proc. 2nd Machine Learning for Healthcare Conference, vol. 68, 322\u2013337 (PMLR, 2017)."},{"key":"2890_CR30","unstructured":"Choi, E. et al. RETAIN: an interpretable predictive model for healthcare using reverse time attention mechanism. In Proc. 30th International Conference on Neural Information Processing Systems 3512\u20133520 (ACM, 2016)."},{"key":"2890_CR31","first-page":"147","volume":"28","author":"NR Marsilio","year":"2016","unstructured":"Marsilio, N. R., da Silva, D. & Bueno, D. Drug incompatibilities in the adult intensive care unit of a university hospital. Rev. Bras. Ter. Intensiv. 28, 147\u2013153 (2016).","journal-title":"Rev. Bras. Ter. Intensiv."},{"key":"2890_CR32","doi-asserted-by":"publisher","DOI":"10.1186\/cc6849","volume":"12","author":"A Delaney","year":"2008","unstructured":"Delaney, A. et al. Bench-to-bedside review: the evaluation of complex interventions in critical care. Crit. Care 12, 210 (2008).","journal-title":"Crit. Care"},{"key":"2890_CR33","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-019-0103-9","volume":"6","author":"H Harutyunyan","year":"2019","unstructured":"Harutyunyan, H., Khachatrian, H., Kale, D. C., Steeg, G., Ver & Galstyan, A. Multitask learning and benchmarking with clinical time series data. Sci. Data 6, 96 (2019).","journal-title":"Sci. Data"},{"key":"2890_CR34","doi-asserted-by":"publisher","first-page":"111196","DOI":"10.1016\/j.compbiomed.2025.111196","volume":"198","author":"I Olier","year":"2025","unstructured":"Olier, I. et al. The integrated multiple event representation framework (IMERF): a case study on critically-ill patients. Comput. Biol. Med. 198, 111196 (2025).","journal-title":"Comput. Biol. Med."},{"key":"2890_CR35","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1023\/A:1007379606734","volume":"28","author":"R Caruana","year":"1997","unstructured":"Caruana, R., Pratt, L. & Thrun, S. Multitask learning. Mach. Learn. 28, 41\u201375 (1997).","journal-title":"Mach. Learn."},{"key":"2890_CR36","doi-asserted-by":"publisher","first-page":"1052","DOI":"10.1093\/jamia\/ocx030","volume":"24","author":"SE Davis","year":"2017","unstructured":"Davis, S. E., Lasko, T. A., Chen, G., Siew, E. D. & Matheny, M. E. Calibration drift in regression and machine learning models for acute kidney injury. J. Am. Med. Inform. Assoc. 24, 1052\u20131061 (2017).","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"2890_CR37","first-page":"346","volume":"44","author":"B Popat","year":"2016","unstructured":"Popat, B. & Jones, A. T. Invasive and non-invasive mechanical ventilation. Medicines 44, 346\u2013350 (2016).","journal-title":"Medicines"},{"key":"2890_CR38","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s13054-022-03960-y","volume":"26","author":"R Suzuki","year":"2022","unstructured":"Suzuki, R., Uchino, S., Sasabuchi, Y., Kawarai Lefor, A. & Sanui, M. Dopamine use and its consequences in the intensive care unit: a cohort study utilizing the Japanese Intensive Care Patient Database. Crit. Care 26, 1\u201312 (2022).","journal-title":"Crit. Care"},{"key":"2890_CR39","doi-asserted-by":"publisher","first-page":"249","DOI":"10.1177\/1074248414559838","volume":"20","author":"JC Jentzer","year":"2015","unstructured":"Jentzer, J. C., Coons, J. C., Link, C. B. & Schmidhofer, M. Pharmacotherapy update on the use of vasopressors and inotropes in the intensive care unit. J. Cardiovasc. Pharmacol. Ther. 20, 249\u2013260 (2015).","journal-title":"J. Cardiovasc. Pharmacol. Ther."},{"key":"2890_CR40","doi-asserted-by":"publisher","first-page":"3188","DOI":"10.3390\/jcm12093188","volume":"12","author":"M Guarino","year":"2023","unstructured":"Guarino, M. et al. 2023 Update on sepsis and septic shock in adult patients: management in the emergency department. J. Clin. Med. 12, 3188 (2023).","journal-title":"J. Clin. Med."},{"key":"2890_CR41","doi-asserted-by":"publisher","first-page":"502","DOI":"10.1097\/CCM.0b013e318232da75","volume":"40","author":"DM Needham","year":"2012","unstructured":"Needham, D. M. et al. Improving long-term outcomes after discharge from intensive care unit: rport from a stakeholders\u2019 conference. Crit. Care Med. 40, 502\u2013509 (2012).","journal-title":"Crit. Care Med."},{"key":"2890_CR42","doi-asserted-by":"publisher","first-page":"202","DOI":"10.1007\/s40140-019-00323-2","volume":"9","author":"D Van Beersel","year":"2019","unstructured":"Van Beersel, D. & Rex, S. Inotropes and vasoactive agents: differences between Europe and the United States. Curr. Anesthesiol. Rep. 9, 202\u2013213 (2019).","journal-title":"Curr. Anesthesiol. Rep."},{"key":"2890_CR43","doi-asserted-by":"publisher","DOI":"10.1186\/s13054-025-05505-5","volume":"29","author":"M Jozwiak","year":"2025","unstructured":"Jozwiak, M. et al. Vasopressin use across shock states: international insights from an international ESICM-endorsed survey: the PRESS Survey. Crit. Care 29, 273 (2025).","journal-title":"Crit. Care"},{"key":"2890_CR44","doi-asserted-by":"publisher","first-page":"621","DOI":"10.1093\/jamia\/ocz228","volume":"27","author":"Y Huang","year":"2020","unstructured":"Huang, Y., Li, W., Macheret, F., Gabriel, R. A. & Ohno-Machado, L. A tutorial on calibration measurements and calibration models for clinical prediction models. J. Am. Med. Inform. Assoc. 27, 621\u2013633 (2020).","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"2890_CR45","doi-asserted-by":"publisher","first-page":"128","DOI":"10.1097\/EDE.0b013e3181c30fb2","volume":"21","author":"EW Steyerberg","year":"2010","unstructured":"Steyerberg, E. W. et al. Assessing the performance of prediction models: a framework for traditional and novel measures. Epidemiology 21, 128\u2013138 (2010).","journal-title":"Epidemiology"},{"key":"2890_CR46","doi-asserted-by":"publisher","first-page":"1936","DOI":"10.1093\/jamia\/ocab101","volume":"28","author":"S Roy","year":"2021","unstructured":"Roy, S. et al. Multitask prediction of organ dysfunction in the intensive care unit using sequential subnetwork routing. J. Am. Med. Inform. Assoc. 28, 1936\u20131946 (2021).","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"2890_CR47","doi-asserted-by":"publisher","first-page":"877","DOI":"10.1056\/NEJMoa067373","volume":"358","author":"JA Russell","year":"2008","unstructured":"Russell, J. A. et al. Vasopressin versus Norepinephrine Infusion in Patients with Septic Shock. N. Engl. J. Med. 358, 877\u2013887 (2008).","journal-title":"N. Engl. J. Med."},{"key":"2890_CR48","doi-asserted-by":"publisher","first-page":"1181","DOI":"10.1007\/s00134-021-06506-y","volume":"47","author":"L Evans","year":"2021","unstructured":"Evans, L. et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock 2021. Intensive Care Med 47, 1181\u20131247 (2021).","journal-title":"Intensive Care Med"},{"key":"2890_CR49","doi-asserted-by":"publisher","first-page":"739","DOI":"10.1007\/s00134-016-4326-3","volume":"42","author":"A Vieillard-Baron","year":"2016","unstructured":"Vieillard-Baron, A. et al. Experts\u2019 opinion on management of hemodynamics in ARDS patients: focus on the effects of mechanical ventilation. Intensive Care Med. 42, 739\u2013749 (2016).","journal-title":"Intensive Care Med."},{"key":"2890_CR50","doi-asserted-by":"publisher","first-page":"488","DOI":"10.1093\/jamia\/ocw138","volume":"24","author":"M Wu","year":"2017","unstructured":"Wu, M. et al. Understanding vasopressor intervention and weaning: risk prediction in a public heterogeneous clinical time series database. J. Am. Med. Inform. Assoc. 24, 488\u2013495 (2017).","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"2890_CR51","unstructured":"Ba, J. L., Kiros, J. R. & Hinton, G. E. Layer normalization. https:\/\/arxiv.org\/pdf\/1607.06450 (2016)."},{"key":"2890_CR52","doi-asserted-by":"publisher","first-page":"847","DOI":"10.1164\/rccm.201006-0972CI","volume":"183","author":"SM Hollenberg","year":"2011","unstructured":"Hollenberg, S. M. Vasoactive drugs in circulatory shock. Am. J. Respir. Crit. Care Med. 183, 847\u2013855 (2011).","journal-title":"Am. J. Respir. Crit. Care Med."},{"key":"2890_CR53","doi-asserted-by":"publisher","first-page":"172","DOI":"10.1177\/0885066610396312","volume":"27","author":"TS Vasu","year":"2012","unstructured":"Vasu, T. S. et al. Norepinephrine or dopamine for septic shock: systematic review of randomized clinical trials. J. Intensive Care Med. 27, 172\u2013178 (2012).","journal-title":"J. Intensive Care Med."},{"key":"2890_CR54","doi-asserted-by":"publisher","first-page":"779","DOI":"10.1056\/NEJMoa0907118","volume":"362","author":"D De Backer","year":"2010","unstructured":"De Backer, D. et al. Comparison of dopamine and norepinephrine in the treatment of shock. N. Engl. J. Med. 362, 779\u2013789 (2010).","journal-title":"N. Engl. J. Med."},{"key":"2890_CR55","doi-asserted-by":"crossref","unstructured":"Scheeren, T. W. L. et al. Current use of vasopressors in septic shock. Ann. Intensive Care 9, 20 (2019).","DOI":"10.1186\/s13613-019-0498-7"},{"key":"2890_CR56","unstructured":"Standley, T. et al. Which tasks should be learned together in multi-task learning? In Proc. 37th International Conference on Machine Learning 9120\u20139132 (ACM,2020)."},{"key":"2890_CR57","unstructured":"Yu, T. et al. Gradient surgery for multi-task learning. In Proc. 34th International Conference on Neural Information Processing Systems (ACM, 2020)."},{"key":"2890_CR58","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0118432","volume":"10","author":"T Saito","year":"2015","unstructured":"Saito, T. & Rehmsmeier, M. The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PLoS ONE 10, e0118432 (2015).","journal-title":"PLoS ONE"},{"key":"2890_CR59","unstructured":"Crawshaw, M. Multi-task learning with deep neural networks: a survey. Preprint at https:\/\/arxiv.org\/pdf\/2009.09796 (2020)."},{"key":"2890_CR60","first-page":"3614","volume":"44","author":"S Vandenhende","year":"2021","unstructured":"Vandenhende, S. et al. Multi-task learning for dense prediction tasks: a survey. IEEE Trans. Pattern Anal. Mach. Intell. 44, 3614\u20133633 (2021).","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"2890_CR61","doi-asserted-by":"publisher","first-page":"274","DOI":"10.1093\/jamia\/ocad178","volume":"31","author":"SE Davis","year":"2023","unstructured":"Davis, S. E., Matheny, M. E., Balu, S. & Sendak, M. P. A framework for understanding label leakage in machine learning for health care. J. Am. Med. Inform. Assoc. 31, 274\u2013280 (2023).","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"2890_CR62","doi-asserted-by":"publisher","first-page":"e2550454","DOI":"10.1001\/jamanetworkopen.2025.50454","volume":"8","author":"B Ramadan","year":"2025","unstructured":"Ramadan, B., Liu, M. C., Burkhart, M. C., Parker, W. F. & Beaulieu-Jones, B. K. Diagnostic codes in AI prediction models and label leakage of same-admission clinical outcomes. JAMA Netw. Open 8, e2550454\u2013e2550454 (2025).","journal-title":"JAMA Netw. Open"},{"key":"2890_CR63","doi-asserted-by":"publisher","first-page":"164","DOI":"10.1126\/science.adg8538","volume":"383","author":"AM Chekroud","year":"2024","unstructured":"Chekroud, A. M. et al. Illusory generalizability of clinical prediction models. Science 383, 164\u2013167 (2024).","journal-title":"Science"},{"key":"2890_CR64","doi-asserted-by":"publisher","first-page":"1347","DOI":"10.1056\/NEJMra1814259","volume":"380","author":"A Rajkomar","year":"2019","unstructured":"Rajkomar, A., Dean, J. & Kohane, I. Machine learning in medicine. N. Engl. J. Med. 380, 1347\u20131358 (2019).","journal-title":"N. Engl. J. Med."},{"key":"2890_CR65","doi-asserted-by":"publisher","first-page":"103611","DOI":"10.1016\/j.jbi.2020.103611","volume":"112","author":"SE Davis","year":"2020","unstructured":"Davis, S. E., Greevy, R. A., Lasko, T. A., Walsh, C. G. & Matheny, M. E. Detection of calibration drift in clinical prediction models to inform model updating. J. Biomed. Inform. 112, 103611 (2020).","journal-title":"J. Biomed. Inform."},{"key":"2890_CR66","doi-asserted-by":"publisher","first-page":"215","DOI":"10.1162\/089976698300017953","volume":"10","author":"CM Bishop","year":"1998","unstructured":"Bishop, C. M., Svens\u00e9n, M. & Williams, C. K. I. GTM: the generative topographic mapping. Neural Comput. 10, 215\u2013234 (1998).","journal-title":"Neural Comput."},{"key":"2890_CR67","doi-asserted-by":"publisher","first-page":"1793","DOI":"10.1097\/00003246-199811000-00016","volume":"26","author":"JL Vincent","year":"1998","unstructured":"Vincent, J. L. et al. Use of the SOFA score to assess the incidence of organ dysfunction\/failure in intensive care units: results of a multicenter, prospective study. Working group on \u2018sepsis-related problems\u2019 of the European Society of Intensive Care Medicine. Crit. Care Med. 26, 1793\u20131800 (1998).","journal-title":"Crit. Care Med."},{"key":"2890_CR68","doi-asserted-by":"publisher","first-page":"156","DOI":"10.1053\/j.jvca.2012.06.014","volume":"27","author":"B Mets","year":"2013","unstructured":"Mets, B. Management of hypotension associated with angiotensin-axis blockade and general anesthesia administration. J. Cardiothorac. Vasc. Anesth. 27, 156\u2013167 (2013).","journal-title":"J. Cardiothorac. Vasc. Anesth."},{"key":"2890_CR69","doi-asserted-by":"publisher","first-page":"1","DOI":"10.3390\/ijerph17228409","volume":"17","author":"K Lewandowska","year":"2020","unstructured":"Lewandowska, K. et al. Impact of alarm fatigue on the work of nurses in an intensive care environment-a systematic review. Int. J. Environ. Res. Public Health 17, 1\u201314 (2020).","journal-title":"Int. J. Environ. Res. Public Health"},{"key":"2890_CR70","doi-asserted-by":"publisher","first-page":"1337","DOI":"10.1038\/s41591-019-0548-6","volume":"25","author":"J Wiens","year":"2019","unstructured":"Wiens, J. et al. Do no harm: a roadmap for responsible machine learning for health care. Nat. Med. 25, 1337\u20131340 (2019).","journal-title":"Nat. Med."},{"key":"2890_CR71","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-023-43715-z","volume":"14","author":"Z Yang","year":"2023","unstructured":"Yang, Z., Mitra, A., Liu, W., Berlowitz, D. & Yu, H. TransformEHR: transformer-based encoder-decoder generative model to enhance prediction of disease outcomes using electronic health records. Nat. Commun. 14, 7857 (2023).","journal-title":"Nat. Commun."},{"key":"2890_CR72","doi-asserted-by":"publisher","first-page":"355","DOI":"10.1093\/jamia\/ocz205","volume":"27","author":"FJR Catling","year":"2020","unstructured":"Catling, F. J. R. & Wolff, A. H. Temporal convolutional networks allow early prediction of events in critical care. J. Am. Med. Inform. Assoc. 27, 355\u2013365 (2020).","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"2890_CR73","doi-asserted-by":"crossref","unstructured":"Oss Boll, H. et al. Graph neural networks for clinical risk prediction based on electronic health records: a survey. J. Biomed. Inform. 151, 104616 (2024).","DOI":"10.1016\/j.jbi.2024.104616"},{"key":"2890_CR74","doi-asserted-by":"publisher","DOI":"10.1038\/s41597-019-0103-9","volume":"6","author":"H Harutyunyan","year":"2019","unstructured":"Harutyunyan, H., Khachatrian, H., Kale, D. C., Ver Steeg, G. & Galstyan, A. Multitask learning and benchmarking with clinical time series data. Sci. Data 6, 96 (2019).","journal-title":"Sci. Data"},{"key":"2890_CR75","unstructured":"Shickel, B., Tighe, P. J., Bihorac, A. & Rashidi, P. Multi-task prediction of clinical outcomes in the intensive care unit using flexible multimodal transformers. https:\/\/arxiv.org\/pdf\/2111.05431 (2021)."},{"key":"2890_CR76","doi-asserted-by":"publisher","first-page":"40","DOI":"10.1002\/mpr.329","volume":"20","author":"MJ Azur","year":"2011","unstructured":"Azur, M. J., Stuart, E. A., Frangakis, C. & Leaf, P. J. Multiple imputation by chained equations: what is it and how does it work? Int. J. Methods Psychiatr. Res. 20, 40\u201349 (2011).","journal-title":"Int. J. Methods Psychiatr. Res."},{"key":"2890_CR77","unstructured":"Lin, J., Lei, H. & Michailidis, G. A VAE-based framework for learning multi-level neural Granger-causal connectivity. Transact. Mach. Learn. Res. Preprint at https:\/\/openreview.net\/forum?id=kNCZ95mw7N (2024)."},{"key":"2890_CR78","unstructured":"Battaglia, P. W. et al. Relational inductive biases, deep learning, and graph networks. https:\/\/arxiv.org\/pdf\/1806.01261 (2018)."},{"key":"2890_CR79","doi-asserted-by":"publisher","first-page":"318","DOI":"10.1109\/TPAMI.2018.2858826","volume":"42","author":"TY Lin","year":"2017","unstructured":"Lin, T. Y., Goyal, P., Girshick, R., He, K. & Dollar, P. Focal loss for dense object detection. IEEE Trans. Pattern Anal. Mach. Intell. 42, 318\u2013327 (2017).","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"2890_CR80","doi-asserted-by":"crossref","unstructured":"Bellfield, R. A. A. et al. AI-based derivation of atrial fibrillation phenotypes in the general and critical care populations. EBioMedicine 107, 105280 (2024).","DOI":"10.1016\/j.ebiom.2024.105280"},{"key":"2890_CR81","doi-asserted-by":"publisher","DOI":"10.1186\/s13040-025-00514-0","volume":"18","author":"J Li","year":"2025","unstructured":"Li, J., Zhao, M. & He, Q. Subphenotype heterogeneity to guide predictive enrichment in acute kidney injury: insights from machine learning and target trial emulation. BioData Min. 18, 87 (2025).","journal-title":"BioData Min."},{"key":"2890_CR82","doi-asserted-by":"crossref","unstructured":"Orlov, A. A., Akhmetshin, T. N., Horvath, D., Marcou, G. & Varnek, A. From high dimensions to human insight: exploring dimensionality reduction for chemical space visualization. Mol. Inform. 44, e202400265 (2025).","DOI":"10.1002\/minf.202400265"},{"key":"2890_CR83","doi-asserted-by":"publisher","first-page":"301","DOI":"10.1002\/minf.201100163","volume":"31","author":"N Kireeva","year":"2012","unstructured":"Kireeva, N. et al. Generative topographic mapping (GTM): universal tool for data visualization, structure-activity modeling and dataset comparison. Mol. Inform. 31, 301\u2013312 (2012).","journal-title":"Mol. Inform."},{"key":"2890_CR84","doi-asserted-by":"publisher","first-page":"99","DOI":"10.1016\/j.ddtec.2020.06.003","volume":"32\u201333","author":"D Horvath","year":"2019","unstructured":"Horvath, D., Marcou, G. & Varnek, A. Generative topographic mapping in drug design. Drug Discov. Today Technol. 32\u201333, 99\u2013107 (2019).","journal-title":"Drug Discov. Today Technol."},{"key":"2890_CR85","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s42452-023-05554-x","volume":"5","author":"S Ortega-Martorell","year":"2023","unstructured":"Ortega-Martorell, S. et al. Mapping the global free expression landscape using machine learning. SN Appl. Sci. 5, 1\u201312 (2023).","journal-title":"SN Appl. Sci."},{"key":"2890_CR86","doi-asserted-by":"crossref","unstructured":"Efron, B. Bootstrap methods: another look at the jackknife. Ann. Stat. 7, 1\u201326 (1979).","DOI":"10.1214\/aos\/1176344552"},{"key":"2890_CR87","doi-asserted-by":"publisher","first-page":"44","DOI":"10.1038\/s41591-018-0300-7","volume":"25","author":"EJ Topol","year":"2019","unstructured":"Topol, E. J. High-performance medicine: the convergence of human and artificial intelligence. Nat. Med. 25, 44\u201356 (2019).","journal-title":"Nat. Med."}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-026-02890-1","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-026-02890-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-026-02890-1.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T00:02:19Z","timestamp":1790208139000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-026-02890-1"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,11]]},"references-count":87,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["2890"],"URL":"https:\/\/doi.org\/10.1038\/s41746-026-02890-1","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,11]]},"assertion":[{"value":"19 January 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"4 June 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 June 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Ethics"}}],"article-number":"722"}}