{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,24]],"date-time":"2026-04-24T16:51:36Z","timestamp":1777049496792,"version":"3.51.4"},"reference-count":37,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2023,8,21]],"date-time":"2023-08-21T00:00:00Z","timestamp":1692576000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2023,8,21]],"date-time":"2023-08-21T00:00:00Z","timestamp":1692576000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100005370","name":"Gates Cambridge Trust","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100005370","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100011272","name":"EC | EC Seventh Framework Programm | FP7 Health","doi-asserted-by":"publisher","award":["602150"],"award-info":[{"award-number":["602150"]}],"id":[{"id":"10.13039\/100011272","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><jats:p>Existing methods to characterise the evolving condition of traumatic brain injury (TBI) patients in the intensive care unit (ICU) do not capture the context necessary for individualising treatment. Here, we integrate all heterogenous data stored in medical records (1166 pre-ICU and ICU variables) to model the individualised contribution of clinical course to 6-month functional outcome on the Glasgow Outcome Scale -Extended (GOSE). On a prospective cohort (<jats:italic>n<\/jats:italic>\u2009=\u20091550, 65 centres) of TBI patients, we train recurrent neural network models to map a token-embedded time series representation of all variables (including missing values) to an ordinal GOSE prognosis every 2\u2009h. The full range of variables explains up to 52% (95% CI: 50\u201354%) of the ordinal variance in functional outcome. Up to 91% (95% CI: 90\u201391%) of this explanation is derived from pre-ICU and admission information (i.e., static variables). Information collected in the ICU (i.e., dynamic variables) increases explanation (by up to 5% [95% CI: 4\u20136%]), though not enough to counter poorer overall performance in longer-stay (&gt;5.75 days) patients. Highest-contributing variables include physician-based prognoses, CT features, and markers of neurological function. Whilst static information currently accounts for the majority of functional outcome explanation after TBI, data-driven analysis highlights investigative avenues to improve the dynamic characterisation of longer-stay patients. Moreover, our modelling strategy proves useful for converting large patient records into interpretable time series with missing data integration and minimal processing.<\/jats:p>","DOI":"10.1038\/s41746-023-00895-8","type":"journal-article","created":{"date-parts":[[2023,8,21]],"date-time":"2023-08-21T08:02:36Z","timestamp":1692604956000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":18,"title":["Mining the contribution of intensive care clinical course to outcome after traumatic brain injury"],"prefix":"10.1038","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7428-5588","authenticated-orcid":false,"given":"Shubhayu","family":"Bhattacharyay","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0839-4396","authenticated-orcid":false,"given":"Pier Francesco","family":"Caruso","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4918-1482","authenticated-orcid":false,"given":"Cecilia","family":"\u00c5kerlund","sequence":"additional","affiliation":[]},{"given":"Lindsay","family":"Wilson","sequence":"additional","affiliation":[]},{"given":"Robert D.","family":"Stevens","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3228-9692","authenticated-orcid":false,"given":"David K.","family":"Menon","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7787-0122","authenticated-orcid":false,"given":"Ewout W.","family":"Steyerberg","sequence":"additional","affiliation":[]},{"given":"David W.","family":"Nelson","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8350-8093","authenticated-orcid":false,"given":"Ari","family":"Ercole","sequence":"additional","affiliation":[]},{"name":"the CENTER-TBI investigators and participants","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2023,8,21]]},"reference":[{"key":"895_CR1","doi-asserted-by":"publisher","first-page":"987","DOI":"10.1016\/S1474-4422(17)30371-X","volume":"16","author":"AIR Maas","year":"2017","unstructured":"Maas, A. I. R. et al. Traumatic brain injury: integrated approaches to improve prevention, clinical care, and research. Lancet Neurol. 16, 987\u20131048 (2017).","journal-title":"Lancet Neurol."},{"key":"895_CR2","doi-asserted-by":"publisher","first-page":"1004","DOI":"10.1016\/S1474-4422(22)00309-X","volume":"21","author":"AIR Maas","year":"2022","unstructured":"Maas, A. I. R. et al. Traumatic brain injury: progress and challenges in prevention, clinical care, and research. Lancet Neurol. 21, 1004\u20131060 (2022).","journal-title":"Lancet Neurol."},{"key":"895_CR3","doi-asserted-by":"publisher","first-page":"649","DOI":"10.1007\/s00134-022-06702-4","volume":"48","author":"G Meyfroidt","year":"2022","unstructured":"Meyfroidt, G. et al. Management of moderate to severe traumatic brain injury: an update for the intensivist. Intensive Care Med. 48, 649\u2013666 (2022).","journal-title":"Intensive Care Med."},{"key":"895_CR4","doi-asserted-by":"publisher","first-page":"844","DOI":"10.1016\/S1474-4422(14)70120-6","volume":"13","author":"G Teasdale","year":"2014","unstructured":"Teasdale, G. et al. The Glasgow Coma Scale at 40 years: standing the test of time. Lancet Neurol. 13, 844\u2013854 (2014).","journal-title":"Lancet Neurol."},{"key":"895_CR5","doi-asserted-by":"publisher","first-page":"923","DOI":"10.1016\/S1474-4422(19)30232-7","volume":"18","author":"EW Steyerberg","year":"2019","unstructured":"Steyerberg, E. W. et al. Case-mix, care pathways, and outcomes in patients with traumatic brain injury in CENTER-TBI: a European prospective, multicentre, longitudinal, cohort study. Lancet Neurol. 18, 923\u2013934 (2019).","journal-title":"Lancet Neurol."},{"key":"895_CR6","doi-asserted-by":"publisher","first-page":"477","DOI":"10.1038\/nrneurol.2016.89","volume":"12","author":"T McMillan","year":"2016","unstructured":"McMillan, T. et al. The Glasgow Outcome Scale\u201440 years of application and refinement. Nat. Rev. Neurol. 12, 477\u2013485 (2016).","journal-title":"Nat. Rev. Neurol."},{"key":"895_CR7","doi-asserted-by":"publisher","first-page":"e165","DOI":"10.1371\/journal.pmed.0050165","volume":"5","author":"EW Steyerberg","year":"2008","unstructured":"Steyerberg, E. W. et al. Predicting outcome after traumatic brain injury: development and international validation of prognostic scores based on admission characteristics. PLoS Med. 5, e165 (2008).","journal-title":"PLoS Med."},{"key":"895_CR8","doi-asserted-by":"publisher","first-page":"e0270973","DOI":"10.1371\/journal.pone.0270973","volume":"17","author":"S Bhattacharyay","year":"2022","unstructured":"Bhattacharyay, S. et al. The leap to ordinal: detailed functional prognosis after traumatic brain injury with a flexible modelling approach. PLoS ONE 17, e0270973 (2022).","journal-title":"PLoS ONE"},{"key":"895_CR9","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1227\/NEU.0000000000000575","volume":"76","author":"AIR Maas","year":"2015","unstructured":"Maas, A. I. R. et al. Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI): a prospective longitudinal observational study. Neurosurgery 76, 67\u201380 (2015).","journal-title":"Neurosurgery"},{"key":"895_CR10","doi-asserted-by":"crossref","unstructured":"Bento, J., Saleiro, P., Cruz, A. F., Figueiredo, M. A. T. & Bizarro, P. TimeSHAP: explaining recurrent models through sequence perturbations. in Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining (KDD \u201921) (eds Zhu, F. et al.) 2565\u20132573 (Association for Computing Machinery, 2021).","DOI":"10.1145\/3447548.3467166"},{"key":"895_CR11","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-79142-z","volume":"10","author":"J Deasy","year":"2020","unstructured":"Deasy, J., Li\u00f2, P. & Ercole, A. Dynamic survival prediction in intensive care units from heterogeneous time series without the need for variable selection or curation. Sci. Rep. 10, 22129 (2020).","journal-title":"Sci. Rep."},{"key":"895_CR12","doi-asserted-by":"publisher","first-page":"e0253425","DOI":"10.1371\/journal.pone.0253425","volume":"16","author":"A Ercole","year":"2021","unstructured":"Ercole, A. et al. Imputation strategies for missing baseline neurological assessment covariates after traumatic brain injury: a CENTER-TBI study. PLoS ONE 16, e0253425 (2021).","journal-title":"PLoS ONE"},{"key":"895_CR13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3477600","volume":"55","author":"C Yu","year":"2021","unstructured":"Yu, C., Liu, J., Nemati, S. & Yin, G. Reinforcement learning in healthcare: a survey. ACM Comput. Surv. 55, 1\u201336 (2021).","journal-title":"ACM Comput. Surv."},{"key":"895_CR14","doi-asserted-by":"publisher","first-page":"87","DOI":"10.1002\/cpt.1907","volume":"109","author":"I Bica","year":"2021","unstructured":"Bica, I., Alaa, A. M., Lambert, C. & van der Schaar, M. From real-world patient data to individualized treatment effects using machine learning: current and future methods to address underlying challenges. Clin. Pharmacol. Ther. 109, 87\u2013100 (2021).","journal-title":"Clin. Pharmacol. Ther."},{"key":"895_CR15","doi-asserted-by":"publisher","first-page":"995","DOI":"10.1007\/s00134-020-05965-z","volume":"46","author":"JA Huijben","year":"2020","unstructured":"Huijben, J. A. et al. Changing care pathways and between-center practice variations in intensive care for traumatic brain injury across Europe: a CENTER-TBI analysis. Intensive Care Med. 46, 995\u20131004 (2020).","journal-title":"Intensive Care Med."},{"key":"895_CR16","doi-asserted-by":"publisher","first-page":"407","DOI":"10.1136\/jnnp-2020-324492","volume":"92","author":"L Wilson","year":"2021","unstructured":"Wilson, L. et al. Understanding the relationship between cognitive performance and function in daily life after traumatic brain injury. J. Neurol. Neurosurg. Psychiatry 92, 407\u2013417 (2021).","journal-title":"J. Neurol. Neurosurg. Psychiatry"},{"key":"895_CR17","doi-asserted-by":"publisher","first-page":"1363","DOI":"10.1089\/neu.2022.0013","volume":"39","author":"L Wilson","year":"2022","unstructured":"Wilson, L. et al. Tailoring multi-dimensional outcomes to level of functional recovery after traumatic brain injury. J. Neurotrauma 39, 1363\u20131381 (2022).","journal-title":"J. Neurotrauma"},{"key":"895_CR18","doi-asserted-by":"publisher","first-page":"101458","DOI":"10.1016\/j.rehab.2020.10.009","volume":"64","author":"IMH Borgen","year":"2021","unstructured":"Borgen, I. M. H. et al. Care transitions in the first 6 months following traumatic brain injury: lessons from the CENTER-TBI study. Ann. Phys. Rehabil. Med. 64, 101458 (2021).","journal-title":"Ann. Phys. Rehabil. Med."},{"key":"895_CR19","doi-asserted-by":"publisher","first-page":"103933","DOI":"10.1016\/j.ebiom.2022.103933","volume":"77","author":"M Kals","year":"2022","unstructured":"Kals, M. et al. A genome-wide association study of outcome from traumatic brain injury. eBioMedicine 77, 103933 (2022).","journal-title":"eBioMedicine"},{"key":"895_CR20","doi-asserted-by":"publisher","first-page":"347","DOI":"10.1007\/s12028-013-9925-z","volume":"19","author":"S Izzy","year":"2013","unstructured":"Izzy, S., Compton, R., Carandang, R., Hall, W. & Muehlschlegel, S. Self-fulfilling prophecies through withdrawal of care: do they exist in traumatic brain injury, too? Neurocrit. Care 19, 347\u2013363 (2013).","journal-title":"Neurocrit. Care"},{"key":"895_CR21","doi-asserted-by":"publisher","first-page":"1115","DOI":"10.1007\/s00134-021-06484-1","volume":"47","author":"E van Veen","year":"2021","unstructured":"van Veen, E. et al. Occurrence and timing of withdrawal of life-sustaining measures in traumatic brain injury patients: a CENTER-TBI study. Intensive Care Med. 47, 1115\u20131129 (2021).","journal-title":"Intensive Care Med."},{"key":"895_CR22","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1038\/s41746-022-00652-3","volume":"5","author":"R Raj","year":"2022","unstructured":"Raj, R. et al. Dynamic prediction of mortality after traumatic brain injury using a machine learning algorithm. NPJ Digit. Med. 5, 96 (2022).","journal-title":"NPJ Digit. Med."},{"key":"895_CR23","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-021-02974-w","volume":"11","author":"S Bhattacharyay","year":"2021","unstructured":"Bhattacharyay, S. et al. Decoding accelerometry for classification and prediction of critically ill patients with severe brain injury. Sci. Rep. 11, 23654 (2021).","journal-title":"Sci. Rep."},{"key":"895_CR24","doi-asserted-by":"publisher","first-page":"419","DOI":"10.1007\/s00134-016-4655-2","volume":"43","author":"MB Skrifvars","year":"2017","unstructured":"Skrifvars, M. B. et al. Venous thromboembolic events in critically ill traumatic brain injury patients. Intensive Care Med. 43, 419\u2013428 (2017).","journal-title":"Intensive Care Med."},{"key":"895_CR25","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-022-01953-z","volume":"22","author":"C Minoccheri","year":"2022","unstructured":"Minoccheri, C. et al. An interpretable neural network for outcome prediction in traumatic brain injury. BMC Med. Inform. Decis. Mak. 22, 203 (2022).","journal-title":"BMC Med. Inform. Decis. Mak."},{"key":"895_CR26","doi-asserted-by":"publisher","first-page":"153","DOI":"10.1016\/S1474-4422(21)00374-4","volume":"21","author":"S Galimberti","year":"2022","unstructured":"Galimberti, S. et al. Effect of frailty on 6-month outcome after traumatic brain injury: a multicentre cohort study with external validation. Lancet Neurol. 21, 153\u2013162 (2022).","journal-title":"Lancet Neurol."},{"key":"895_CR27","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-019-53889-6","volume":"9","author":"R Raj","year":"2019","unstructured":"Raj, R. et al. Machine learning-based dynamic mortality prediction after traumatic brain injury. Sci. Rep. 9, 17672 (2019).","journal-title":"Sci. Rep."},{"key":"895_CR28","first-page":"900","volume":"2021","author":"A Nayebi","year":"2021","unstructured":"Nayebi, A. et al. Recurrent neural network based time-series modeling for long-term prognosis following acute traumatic brain injury. AMIA Annu. Symp. Proc. 2021, 900\u2013909 (2021).","journal-title":"AMIA Annu. Symp. Proc."},{"key":"895_CR29","unstructured":"Kidger, P. On Neural Differential Equations. PhD thesis, University of Oxford (2021)."},{"key":"895_CR30","doi-asserted-by":"publisher","first-page":"e489","DOI":"10.1016\/S2589-7500(20)30186-2","volume":"2","author":"J Futoma","year":"2020","unstructured":"Futoma, J., Simons, M., Panch, T., Doshi-Velez, F. & Celi, L. A. The myth of generalisability in clinical research and machine learning in health care. Lancet Digit. Health 2, e489\u2013e492 (2020).","journal-title":"Lancet Digit. Health"},{"key":"895_CR31","doi-asserted-by":"publisher","first-page":"438","DOI":"10.1016\/S1474-4422(22)00037-0","volume":"21","author":"D Clark","year":"2022","unstructured":"Clark, D. et al. Casemix, management, and mortality of patients receiving emergency neurosurgery for traumatic brain injury in the Global Neurotrauma Outcomes Study: a prospective observational cohort study. Lancet Neurol. 21, 438\u2013449 (2022).","journal-title":"Lancet Neurol."},{"key":"895_CR32","doi-asserted-by":"publisher","first-page":"1372","DOI":"10.1093\/ije\/dyx180","volume":"46","author":"D Doiron","year":"2017","unstructured":"Doiron, D., Marcon, Y., Fortier, I., Burton, P. & Ferretti, V. Software Application Profile: Opal and Mica: open-source software solutions for epidemiological data management, harmonization and dissemination. Int. J. Epidemiol. 46, 1372\u20131378 (2017).","journal-title":"Int. J. Epidemiol."},{"key":"895_CR33","doi-asserted-by":"publisher","first-page":"455","DOI":"10.1089\/neu.2019.6858","volume":"38","author":"K Kunzmann","year":"2021","unstructured":"Kunzmann, K. et al. Imputation of ordinal outcomes: a comparison of approaches in traumatic brain injury. J. Neurotrauma 38, 455\u2013463 (2021).","journal-title":"J. Neurotrauma"},{"key":"895_CR34","doi-asserted-by":"publisher","first-page":"1895","DOI":"10.1007\/s10994-018-5714-4","volume":"107","author":"I Tsamardinos","year":"2018","unstructured":"Tsamardinos, I., Greasidou, E. & Borboudakis, G. Bootstrapping the out-of-sample predictions for efficient and accurate cross-validation. Mach. Learn. 107, 1895\u20131922 (2018).","journal-title":"Mach. Learn."},{"key":"895_CR35","doi-asserted-by":"publisher","first-page":"167","DOI":"10.1016\/j.jclinepi.2015.12.005","volume":"74","author":"B Van Calster","year":"2016","unstructured":"Van Calster, B. et al. A calibration hierarchy for risk models was defined: from utopia to empirical data. J. Clin. Epidemiol. 74, 167\u2013176 (2016).","journal-title":"J. Clin. Epidemiol."},{"key":"895_CR36","doi-asserted-by":"publisher","first-page":"799","DOI":"10.2307\/2090408","volume":"27","author":"RH Somers","year":"1962","unstructured":"Somers, R. H. A new asymmetric measure of association for ordinal variables. Am. Sociol. Rev. 27, 799\u2013811 (1962).","journal-title":"Am. Sociol. Rev."},{"key":"895_CR37","doi-asserted-by":"publisher","first-page":"674","DOI":"10.1002\/bimj.201200026","volume":"54","author":"B Van Calster","year":"2012","unstructured":"Van Calster, B., Van Belle, V., Vergouwe, Y. & Steyerberg, E. W. Discrimination ability of prediction models for ordinal outcomes: relationships between existing measures and a new measure. Biom. J. 54, 674\u2013685 (2012).","journal-title":"Biom. J."}],"container-title":["npj Digital Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00895-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00895-8","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00895-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,11,19]],"date-time":"2023-11-19T17:20:56Z","timestamp":1700414456000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41746-023-00895-8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,8,21]]},"references-count":37,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2023,12]]}},"alternative-id":["895"],"URL":"https:\/\/doi.org\/10.1038\/s41746-023-00895-8","relation":{},"ISSN":["2398-6352"],"issn-type":[{"value":"2398-6352","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,8,21]]},"assertion":[{"value":"9 March 2023","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"1 August 2023","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"21 August 2023","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","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"154"}}