Outcome Prediction for Patients with Traumatic Brain Injury with Dynamic Features from Intracranial Pressure and Arterial Blood Pressure Signals: A Gaussian Process Approach

  • Marco A. F. Pimentel
  • Thomas Brennan
  • Li-wei Lehman
  • Nicolas Kon Kam King
  • Beng-Ti Ang
  • Mengling Feng
Part of the Acta Neurochirurgica Supplement book series (NEUROCHIRURGICA, volume 122)


Previous work has been demonstrated that tracking features describing the dynamic and time-varying patterns in brain monitoring signals provide additional predictive information beyond that derived from static features based on snapshot measurements. To achieve more accurate predictions of outcomes of patients with traumatic brain injury (TBI), we proposed a statistical framework to extract dynamic features from brain monitoring signals based on the framework of Gaussian processes (GPs). GPs provide an explicit probabilistic, nonparametric Bayesian approach to metric regression problems. This not only provides probabilistic predictions, but also gives the ability to cope with missing data and infer model parameters such as those that control the function’s shape, noise level and dynamics of the signal. Through experimental evaluation, we have demonstrated that dynamic features extracted from GPs provide additional predictive information in addition to the features based on the pressure reactivity index (PRx). Significant improvements in patient outcome prediction were achieved by combining GP-based and PRx-based dynamic features. In particular, compared with the a baseline PRx-based model, the combined model achieved over 30 % improvement in prediction accuracy and sensitivity and over 20 % improvement in specificity and the area under the receiver operating characteristic curve.


Gaussian process Intracranial pressure Dynamic features and outcome prediction 



Dr Mengling Feng’s fellowship is funded by A*STAR Graduate Scholarship (AGS). Marco A F Pimentel is supported by the RCUK Digital Economy Program grant number EP/G036861/1 (Oxford Centre for Doctoral Training in Healthcare Innovation) and FCT – Fundação para a Ciência e a Tecnologia – under grant SFRH/DB/79799/2011.

Conflict of Interest Statement

We declare that we have no conflicts of interest.


  1. 1.
    Abad N, Druzgalski C (2009) Medical and engineering support and needs in neurological intensive care units. In: Proceedings of the PAHCE'09, Mexico City, Mexico, pp 157–159Google Scholar
  2. 2.
    Arindam C, Prasanth BN, Andy JK (2002) A data parallel approach for large-scale Gaussian process modeling. Proceeding of SDMGoogle Scholar
  3. 3.
    Balestreri M, Czosnyka M, Steiner LA, Hiler M, Schmidt EA, Matta B, Menon D, Hutchinson P, Pickard JD (2005) Association between outcome, cerebral pressure reactivity and slow ICP waves following head injury. Acta Neurochir Suppl 95:25–28CrossRefPubMedGoogle Scholar
  4. 4.
    Catherine JK, Robert LB, Pamela HM (2008) Intracranial pressure variability and long-term outcome following traumatic brain injury. Acta Neurochir Suppl 102:105–108CrossRefGoogle Scholar
  5. 5.
    Clifton L, Clifton DA, Pimentel MA, Watkinson PJ, Tarassenko L (2013) Gaussian processes for personalized e-health monitoring with wearable sensors. IEEE Trans Biomed Eng 60(1):193–197CrossRefPubMedGoogle Scholar
  6. 6.
    Costa M, Goldberger AL, Peng CK (2002) Multiscale entropy analysis of complex physiologic time series. Phys Rev Lett 89:68–102CrossRefGoogle Scholar
  7. 7.
    Czosnyka M, Pickard JD (2004) Monitoring and interpretation of intracranial pressure. J Neurol Neurosurg Psychiatry 75:813–821CrossRefPubMedPubMedCentralGoogle Scholar
  8. 8.
    Czosnyka M, Hutchinson PJ, Balestreri M, Hiler M, Smielewski P, Pickard JD (2006) Monitoring and interpretation of intracranial pressure after head injury. Acta Neurochir Suppl 96(4):121–125Google Scholar
  9. 9.
    Dutton R, McCunn M (2003) Traumatic brain injury. Curr Opin Crit Care 9(6):503–509CrossRefPubMedGoogle Scholar
  10. 10.
    Edward S, Zoubin G (2005) Sparse Gaussian processes using pseudo-inputs. NIPSGoogle Scholar
  11. 11.
    Feng M, Loy LY, Zhang F, Guan CT et al.(2011) Artifact removal for intracranial pressure monitoring Signals. In: A Robust Solution with Signal Decomposition. Proceedings of 33rd international conference of the IEEE Engineering in Medicine and Biology Society (EMBC)Google Scholar
  12. 12.
    Guendling K, Smielewski P, Czosnyka M, Lewis P, Nortje J, Timofeev I, Hutchinson PJ, Pickard JD (2006) Use of ICM+ software for on-line analysis of intracranial and arterial pressures in head-injured patients. Acta Neurochir Suppl 96(4):108–113CrossRefPubMedGoogle Scholar
  13. 13.
    Hlatky R, Valadka AB, Robertson CS (2005) Intracranial pressure response to induced hypertension: role of dynamic pressure autoregulation. Neurosurgery 57:9117–9123CrossRefGoogle Scholar
  14. 14.
  15. 15.
    Hu X, Xu P, Scalzo F, Vespa P, Bergsneider M (2009) Morphological clustering and analysis of continuous intracranial pressure. IEEE Transactions on Biomedical Engineering 56(3):696–705CrossRefPubMedPubMedCentralGoogle Scholar
  16. 16.
    Hu X, Asgari XP, Vespa P, Bergsneider M (2010) Forecasting ICP elevation based on prescient changes of intracranial pressure waveform morphology. IEEE Transactions on Biomedical Engineering 57(5):1070–1078CrossRefPubMedPubMedCentralGoogle Scholar
  17. 17.
    Ivanov PC, Amaral LA, Goldberger AL, Havlin S, Rosenblum MG, Struzik ZR, Stanley HE (1999) Multifractality in human heartbeat dynamics. Nature 399:461–465CrossRefPubMedGoogle Scholar
  18. 18.
    Jones P, Andrews P, Midgley S, Anderson S, Piper I, Tocher J, Housley A, Corrie J, Slattery J, Dearden N (1994) Measuring the burden of secondary insults in head-injured patients during intensive care. J Neurosurg Anesthesiol 6(1):4–14CrossRefPubMedGoogle Scholar
  19. 19.
    Lee KK, Seow WT, Ng I (2006) Demographical profiles of adult severe traumatic brain injury patients: implications for healthcare planning. Singapore Med J 47(1):31–36PubMedGoogle Scholar
  20. 20.
    Marmarou A, Anderson JRL, Ward JD, Choi SC, Young HF, Eisenberg HM, Foulkes MA, Marshall LF, Jane JA (1991) Impact of ICP instability and hypotension on outcome in patients with severe heal trauma. J Neurosurg 75:59–66Google Scholar
  21. 21.
    Marmarou A (2000) Increased intracranial pressure in head injury and influence of blood volume. J Neurotrauma 9:327–333Google Scholar
  22. 22.
    Rasmussen E, Williams CKI (2006) Gaussian processes for machine learning. MIT Press, ISBN 0-262-18253, usGoogle Scholar
  23. 23.
    Roberts S, Osborne M, Ebden M, Reece S, Gibson N, Aigrain S (2012) Gaussian processes for time-series modelling. Philosophical Transactions of the Royal Society (Part A), 371(1984): onlineGoogle Scholar
  24. 24.
    Robertson C, Valadka A, Hannay H, Contant C, Gopinath S, Cormio M, Uzura M, Grossman R (1999) Prevention of secondary ischemic insults after severe head injury. Crit Care Med 27(10):2086–2095CrossRefPubMedGoogle Scholar
  25. 25.
    Ross N, Eynon CA (2005) Intracranial pressure monitoring. Curr Anaesth Crit Care 16(4):255–261CrossRefGoogle Scholar
  26. 26.
    Signorini D, Andrews P, Jones P, Wardlaw J, Miller J (1999) Predicting survival using simple clinical variables: a case study in traumatic brain injury. J Neurol Neurosurg Psychiatry 66(1):20–25CrossRefPubMedPubMedCentralGoogle Scholar
  27. 27.
    Swiercz M, Mariak Z, Krejza J, Lewko J, Szydlik P (2000) Intracranial pressure processing with artificial neural networks: prediction of ICP trends. Acta Neurochir 142(4):531–542CrossRefGoogle Scholar
  28. 28.
    Wald S, Shackford S (1993) The effect of secondary insults on mortality and long-term disability after severe head injury in a rural region without a trauma system. J Trauma 34(3):377–382CrossRefPubMedGoogle Scholar
  29. 29.
    Wright W (2007) Multimodal monitoring in the ICU: when could it be useful? J Neurol Sci 261(1–2):10–15CrossRefPubMedGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Marco A. F. Pimentel
    • 1
  • Thomas Brennan
    • 2
  • Li-wei Lehman
    • 2
  • Nicolas Kon Kam King
    • 3
  • Beng-Ti Ang
    • 3
  • Mengling Feng
    • 2
    • 4
  1. 1.Department of Engineering ScienceInstitute of Biomedical Engineering, University of OxfordOxfordUK
  2. 2.Laboratory for Computational PhysiologyInstitute for Medical Engineering and Science, E25-505 Massachusetts Institute of TechnologyCambridgeUSA
  3. 3.National Neuroscience InstituteSingaporeSingapore
  4. 4.Department of Data AnalyticsA*STAR, Institute for Infocomm ResearchSingaporeSingapore

Personalised recommendations