A traumatic brain injury prognostic model to support in-hospital triage in a low-income country: a machine learning–based approach

Restricted access

OBJECTIVE

Traumatic brain injury (TBI) is a leading cause of death and disability worldwide, with a disproportionate burden of this injury on low- and middle-income countries (LMICs). Limited access to diagnostic technologies and highly skilled providers combined with high patient volumes contributes to poor outcomes in LMICs. Prognostic modeling as a clinical decision support tool, in theory, could optimize the use of existing resources and support timely treatment decisions in LMICs. The objective of this study was to develop a machine learning–based prognostic model using data from Kilimanjaro Christian Medical Centre in Moshi, Tanzania.

METHODS

This study is a secondary analysis of a TBI data registry including 3138 patients. The authors tested nine different machine learning techniques to identify the prognostic model with the greatest area under the receiver operating characteristic curve (AUC). Input data included demographics, vital signs, injury type, and treatment received. The outcome variable was the discharge score on the Glasgow Outcome Scale–Extended.

RESULTS

The AUC for the prognostic models varied from 66.2% (k-nearest neighbors) to 86.5% (Bayesian generalized linear model). An increasing Glasgow Coma Scale score, increasing pulse oximetry values, and undergoing TBI surgery were predictive of a good recovery, while injuries suffered from a motor vehicle crash and increasing age were predictive of a poor recovery.

CONCLUSIONS

The authors developed a TBI prognostic model with a substantial level of accuracy in a low-resource setting. Further research is needed to externally validate the model and test the algorithm as a clinical decision support tool.

ABBREVIATIONS AUC = area under the receiver operating characteristic curve; ED = emergency department; GCS = Glasgow Coma Scale; GOSE = Glasgow Outcome Scale–Extended; HIC = high-income country; ICU = intensive care unit; KCMC = Kilimanjaro Christian Medical Centre; LMICs = low- and middle-income countries; ML = machine learning; MOI = mechanism of injury; PERRLA = pupils equal, round, and reactive to light and accommodation; TBI = traumatic brain injury.

Article Information

Correspondence Cyrus Elahi: Duke University, Durham, NC. cyrusweal@gmail.com.

INCLUDE WHEN CITING Published online May 10, 2019; DOI: 10.3171/2019.2.JNS182098.

Disclosures Dr. Staton acknowledges salary support funding from the Fogarty International Center (K01, TW010000-01A1), Mr. Rocha acknowledges scholarship support funding from the CAPES foundation (88881.131508/2016-01), and Dr. Haglund acknowledges support from NuVasive and LifeNet for non–study-related clinical or research effort.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Comparison of ML models applied to predict TBI outcome. Each line represents the receiver operating characteristic (ROC) curve for a different ML technique. The Bayesian generalized linear model (Bay. Gen. Lin. Mod) was the top performing technique. Grad. boost. machine = gradient boosting machine; Bay. Add. Reg. Tr. = Bayesian additive regression trees. Figure is available in color online only.

  • View in gallery

    Bar chart for the extracted beta coefficients (Coeff) from the top performing prognostic model, the Bayesian generalized linear model. Orange lines represent variables associated with a poor recovery, whereas blue lines represent variables associated with a good recovery. BP = blood pressure; intercept = value in regression model, included for the sake of transparency. Figure is available in color online only.

  • View in gallery

    The web-based application (A) and the mobile-based application (B). Figure is available in color online only

References

  • 1

    Carpenter JBithell J: Bootstrap confidence intervals: when, which, what? A practical guide for medical statisticians. Stat Med 19:114111642000

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Chawla NVBowyer KWHall LOKegelmeyer WP: SMOTE: synthetic minority over-sampling technique. J Artif Intell Res 16:3213572002

  • 3

    Collins GSReitsma JBAltman DGMoons KG: Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): the TRIPOD Statement. BMC Med 13:12015

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    DeLong ERDeLong DMClarke-Pearson DL: Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics 44:8378451988

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 5

    Dewan MCRattani AGupta SBaticulon REHung YCPunchak M: Estimating the global incidence of traumatic brain injury. J Neurosurg [epub ahead of print April 1 2018. DOI: 10.3171/2017.10.JNS17352]

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6

    Eftekhar BMohammad KArdebili HEGhodsi MKetabchi E: Comparison of artificial neural network and logistic regression models for prediction of mortality in head trauma based on initial clinical data. BMC Med Inform Decis Mak 5:32005

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Fawcett T: An introduction to ROC analysis. Pattern Recognit Lett 27:8618742006

  • 8

    Frattalone ARLing GSF: Moderate and severe traumatic brain injury: pathophysiology and management. Neurosurg Clin N Am 24:3093192013

  • 9

    Haselsberger KPucher RAuer LM: Prognosis after acute subdural or epidural haemorrhage. Acta Neurochir (Wien) 90:1111161988

  • 10

    Hastie TTibshirani RFriedman J: The Elements of Statistical Learninged 2. Berlin: Springer2001

  • 11

    Holmes DEJain LCBeaulieu-Jones B: Machine Learning for Structured Clinical Data. Cham, Switzerland: Springer International Publishing2018

    • Search Google Scholar
    • Export Citation
  • 12

    Hsu MHLi YCChiu WTYen JC: Outcome prediction after moderate and severe head injury using an artificial neural network. Stud Health Technol Inform 116:2412452005

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Jennett BSnoek JBond MRBrooks N: Disability after severe head injury: observations on the use of the Glasgow Outcome Scale. J Neurol Neurosurg Psychiatry 44:2852931981

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Kairalla JACoffey CSMuller KE: GLUMIP 2.0: SAS/IML software for planning internal pilots. J Stat Softw 28:12008

  • 15

    Kalpakis KYang SHu PFMackenzie CFStansbury LGStein DM: Permutation entropy analysis of vital signs data for outcome prediction of patients with severe traumatic brain injury. Comput Biol Med 56:1671742015

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 16

    Kuo BJVaca SDVissoci JRNStaton CAXu LMuhumuza M: A prospective neurosurgical registry evaluating the clinical care of traumatic brain injury patients presenting to Mulago National Referral Hospital in Uganda. PLoS One 12:e01822852017

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Lai YCChen FGGoh MHKoh KF: Predictors of long-term outcome in severe head injury. Ann Acad Med Singapore 27:3263311998

  • 18

    Luo WPhung DTran TGupta SRana SKarmakar C: Guidelines for developing and reporting machine learning predictive models in biomedical research: a multidisciplinary view. J Med Internet Res 18:e3232016

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19

    Mukherjee KKSharma BSRamanathan SMKhandelwal NKak VK: A mathematical outcome prediction model in severe head injury: a pilot study. Neurol India 48:43482000

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Murray CJL: The Global Burden of Disease: A Comprehensive Assessment of Mortality and Disability from Diseases Injuries and Risk Factors in 1990 and Projected to 2020: Summary. Cambridge, MA: Harvard School of Public Health1996

    • Search Google Scholar
    • Export Citation
  • 21

    Murray LSTeasdale GMMurray GDJennett BMiller JDPickard JD: Does prediction of outcome alter patient management? Lancet 341:148714911993

  • 22

    Obermeyer ZEmanuel EJ: Predicting the future—big data, machine learning, and clinical medicine. N Engl J Med 375:121612192016

  • 23

    Perel PArango MClayton TEdwards PKomolafe EPoccock S: Predicting outcome after traumatic brain injury: practical prognostic models based on large cohort of international patients. BMJ 336:4254292008

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Perel PEdwards PWentz RRoberts I: Systematic review of prognostic models in traumatic brain injury. BMC Med Inform Decis Mak 6:382006

  • 25

    Perel PWasserberg JRavi RRShakur HEdwards PRoberts I: Prognosis following head injury: a survey of doctors from developing and developed countries. J Eval Clin Pract 13:4644652007

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26

    Pillai SVKolluri VRPraharaj SS: Outcome prediction model for severe diffuse brain injuries: development and evaluation. Neurol India 51:3453492003

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27

    Ratanalert SChompikul JHirunpat SPheunpathom N: Prognosis of severe head injury: an experience in Thailand. Br J Neurosurg 16:4874932002

  • 28

    Raykar NPYorlets RRLiu CGreenberg SLMKotagal MGoldman R: A qualitative study exploring contextual challenges to surgical care provision in 21 LMICs. Lancet 385 (Suppl 2):S152015

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 29

    Reddy CKAggarwal CC: Healthcare Data Analytics. Boca Raton, FL: CRC Press2015

  • 30

    Robin XTurck NHainard ATiberti NLisacek FSanchez JC: pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics 12:772011

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 31

    Sadaka FPatel DLakshmanan R: The FOUR score predicts outcome in patients after traumatic brain injury. Neurocrit Care 16:951012012

  • 32

    Seelig JMBecker DPMiller JDGreenberg RPWard JDChoi SC: Traumatic acute subdural hematoma: major mortality reduction in comatose patients treated within four hours. N Engl J Med 304:151115181981

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 33

    Senders JTStaples PCKarhade AVZaki MMGormley WBBroekman MLD: Machine learning and neurosurgical outcome prediction: a systematic review. World Neurosurg 109:476486 486.e12018

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 34

    Shi HYHwang SLLee KTLin CL: In-hospital mortality after traumatic brain injury surgery: a nationwide population-based comparison of mortality predictors used in artificial neural network and logistic regression models. J Neurosurg 118:7467522013

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 35

    Staton CAMsilanga DKiwango GVissoci JRde Andrade LLester R: A prospective registry evaluating the epidemiology and clinical care of traumatic brain injury patients presenting to a regional referral hospital in Moshi, Tanzania: challenges and the way forward. Int J Inj Contr Saf Promot 24:69772017

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 36

    Steyerberg EWMushkudiani NPerel PButcher ILu JMcHugh GS: Predicting outcome after traumatic brain injury: development and international validation of prognostic scores based on admission characteristics. PLoS Med 5:e1652008

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 37

    Subaiya SRoberts IKomolafe EPerel P: Predicting intracranial hemorrhage after traumatic brain injury in low and middle-income countries: a prognostic model based on a large, multi-center, international cohort. BMC Emerg Med 12:172012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 38

    Vaca SDKuo BJNickenig Vissoci JRStaton CAXu LWMuhumuza M: Temporal delays along the neurosurgical care continuum for traumatic brain injury patients at a tertiary care hospital in Kampala, Uganda. Neurosurgery 84:951032019

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 39

    van Buuren SGroothuis-Oudshoorn K: Multivariate imputation by chained equations in R. J Stat Softw 45:1672011

  • 40

    White IRRoyston PWood AM: Multiple imputation using chained equations: issues and guidance for practice. Stat Med 30:3773992011

  • 41

    Zador ZSperrin MKing AT: Predictors of outcome in traumatic brain injury: new insight using receiver operating curve indices and Bayesian network analysis. PLoS One 11:e01587622016

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 824 824 134
Full Text Views 85 85 28
PDF Downloads 47 47 15
EPUB Downloads 0 0 0

PubMed

Google Scholar