Mining the contribution of intensive care clinical course to outcome after traumatic brain injury.
Journal
NPJ digital medicine
ISSN: 2398-6352
Titre abrégé: NPJ Digit Med
Pays: England
ID NLM: 101731738
Informations de publication
Date de publication:
21 Aug 2023
21 Aug 2023
Historique:
received:
09
03
2023
accepted:
01
08
2023
medline:
22
8
2023
pubmed:
22
8
2023
entrez:
21
8
2023
Statut:
epublish
Résumé
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 (n = 1550, 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 h. The full range of variables explains up to 52% (95% CI: 50-54%) of the ordinal variance in functional outcome. Up to 91% (95% CI: 90-91%) 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: 4-6%]), though not enough to counter poorer overall performance in longer-stay (>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.
Identifiants
pubmed: 37604980
doi: 10.1038/s41746-023-00895-8
pii: 10.1038/s41746-023-00895-8
pmc: PMC10442346
doi:
Types de publication
Journal Article
Langues
eng
Pagination
154Subventions
Organisme : EC | EC Seventh Framework Programm | FP7 Health (FP7-HEALTH - Specific Programme "Cooperation": Health)
ID : 602150
Informations de copyright
© 2023. Springer Nature Limited.
Références
Maas, A. I. R. et al. Traumatic brain injury: integrated approaches to improve prevention, clinical care, and research. Lancet Neurol. 16, 987–1048 (2017).
doi: 10.1016/S1474-4422(17)30371-X
pubmed: 29122524
Maas, A. I. R. et al. Traumatic brain injury: progress and challenges in prevention, clinical care, and research. Lancet Neurol. 21, 1004–1060 (2022).
doi: 10.1016/S1474-4422(22)00309-X
pubmed: 36183712
pmcid: 10427240
Meyfroidt, G. et al. Management of moderate to severe traumatic brain injury: an update for the intensivist. Intensive Care Med. 48, 649–666 (2022).
doi: 10.1007/s00134-022-06702-4
pubmed: 35595999
Teasdale, G. et al. The Glasgow Coma Scale at 40 years: standing the test of time. Lancet Neurol. 13, 844–854 (2014).
doi: 10.1016/S1474-4422(14)70120-6
pubmed: 25030516
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, 923–934 (2019).
doi: 10.1016/S1474-4422(19)30232-7
pubmed: 31526754
McMillan, T. et al. The Glasgow Outcome Scale—40 years of application and refinement. Nat. Rev. Neurol. 12, 477–485 (2016).
doi: 10.1038/nrneurol.2016.89
pubmed: 27418377
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).
doi: 10.1371/journal.pmed.0050165
pubmed: 18684008
pmcid: 2494563
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).
doi: 10.1371/journal.pone.0270973
pubmed: 35788768
pmcid: 9255749
Maas, A. I. R. et al. Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI): a prospective longitudinal observational study. Neurosurgery 76, 67–80 (2015).
doi: 10.1227/NEU.0000000000000575
pubmed: 25525693
Bento, J., Saleiro, P., Cruz, A. F., Figueiredo, M. A. T. & Bizarro, P. TimeSHAP: explaining recurrent models through sequence perturbations. in Proceedings of the 27
Deasy, J., Liò, 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).
doi: 10.1038/s41598-020-79142-z
pubmed: 33335183
pmcid: 7747558
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).
doi: 10.1371/journal.pone.0253425
pubmed: 34358231
pmcid: 8345855
Yu, C., Liu, J., Nemati, S. & Yin, G. Reinforcement learning in healthcare: a survey. ACM Comput. Surv. 55, 1–36 (2021).
doi: 10.1145/3477600
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, 87–100 (2021).
doi: 10.1002/cpt.1907
pubmed: 32449163
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, 995–1004 (2020).
doi: 10.1007/s00134-020-05965-z
pubmed: 32100061
pmcid: 7210239
Wilson, L. et al. Understanding the relationship between cognitive performance and function in daily life after traumatic brain injury. J. Neurol. Neurosurg. Psychiatry 92, 407–417 (2021).
doi: 10.1136/jnnp-2020-324492
Wilson, L. et al. Tailoring multi-dimensional outcomes to level of functional recovery after traumatic brain injury. J. Neurotrauma 39, 1363–1381 (2022).
doi: 10.1089/neu.2022.0013
pubmed: 35607855
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).
doi: 10.1016/j.rehab.2020.10.009
pubmed: 33246186
Kals, M. et al. A genome-wide association study of outcome from traumatic brain injury. eBioMedicine 77, 103933 (2022).
doi: 10.1016/j.ebiom.2022.103933
pubmed: 35301180
pmcid: 8927841
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, 347–363 (2013).
doi: 10.1007/s12028-013-9925-z
pubmed: 24132565
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, 1115–1129 (2021).
doi: 10.1007/s00134-021-06484-1
pubmed: 34351445
pmcid: 8486724
Raj, R. et al. Dynamic prediction of mortality after traumatic brain injury using a machine learning algorithm. NPJ Digit. Med. 5, 96 (2022).
doi: 10.1038/s41746-022-00652-3
pubmed: 35851612
pmcid: 9293936
Bhattacharyay, S. et al. Decoding accelerometry for classification and prediction of critically ill patients with severe brain injury. Sci. Rep. 11, 23654 (2021).
doi: 10.1038/s41598-021-02974-w
pubmed: 34880296
pmcid: 8654973
Skrifvars, M. B. et al. Venous thromboembolic events in critically ill traumatic brain injury patients. Intensive Care Med. 43, 419–428 (2017).
doi: 10.1007/s00134-016-4655-2
pubmed: 28028552
Minoccheri, C. et al. An interpretable neural network for outcome prediction in traumatic brain injury. BMC Med. Inform. Decis. Mak. 22, 203 (2022).
doi: 10.1186/s12911-022-01953-z
pubmed: 35915430
pmcid: 9341077
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, 153–162 (2022).
doi: 10.1016/S1474-4422(21)00374-4
pubmed: 35065038
Raj, R. et al. Machine learning-based dynamic mortality prediction after traumatic brain injury. Sci. Rep. 9, 17672 (2019).
doi: 10.1038/s41598-019-53889-6
pubmed: 31776366
pmcid: 6881446
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, 900–909 (2021).
pubmed: 35309007
Kidger, P. On Neural Differential Equations. PhD thesis, University of Oxford (2021).
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–e492 (2020).
doi: 10.1016/S2589-7500(20)30186-2
pubmed: 32864600
pmcid: 7444947
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, 438–449 (2022).
doi: 10.1016/S1474-4422(22)00037-0
pubmed: 35305318
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, 1372–1378 (2017).
doi: 10.1093/ije/dyx180
pubmed: 29025122
pmcid: 5837212
Kunzmann, K. et al. Imputation of ordinal outcomes: a comparison of approaches in traumatic brain injury. J. Neurotrauma 38, 455–463 (2021).
doi: 10.1089/neu.2019.6858
pubmed: 33108942
pmcid: 7875604
Tsamardinos, I., Greasidou, E. & Borboudakis, G. Bootstrapping the out-of-sample predictions for efficient and accurate cross-validation. Mach. Learn. 107, 1895–1922 (2018).
doi: 10.1007/s10994-018-5714-4
pubmed: 30393425
pmcid: 6191021
Van Calster, B. et al. A calibration hierarchy for risk models was defined: from utopia to empirical data. J. Clin. Epidemiol. 74, 167–176 (2016).
doi: 10.1016/j.jclinepi.2015.12.005
pubmed: 26772608
Somers, R. H. A new asymmetric measure of association for ordinal variables. Am. Sociol. Rev. 27, 799–811 (1962).
doi: 10.2307/2090408
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, 674–685 (2012).
doi: 10.1002/bimj.201200026
pubmed: 22711459