Outcome Prediction for Patients with Severe Traumatic Brain Injury Using Permutation Entropy Analysis of Electronic Vital Signs Data
Permutation entropy is computationally efficient, robust to noise, and effective to measure complexity. We used this technique to quantify the complexity of continuous vital signs recorded from patients with traumatic brain injury (TBI). Using permutation entropy calculated from early vital signs (initial 10~20% of patient hospital stay time), we built classifiers to predict in-hospital mortality, and mobility measured by 3-month Extended Glasgow Outcome Score (GOSE). Sixty patients with severe TBI produced a skewed dataset that we evaluated for accuracy, sensitivity and specificity. With early vital signs data, the overall prediction accuracy achieved 91.67% for mortality, and 76.67% for 3-month GOSE in testing datasets, using the leave-one-out cross validation. We also applied Receiver Operating Characteristic analysis to compare classifiers built from different learning methods. Those results support the applicability of permutation entropy in analyzing the dynamic behavior of biomedical time series for early prediction of mortality and long-term patient outcomes.
KeywordsTraumatic Brain Injury Vital Sign Cerebral Perfusion Pressure Severe Traumatic Brain Injury Permutation Entropy
Unable to display preview. Download preview PDF.
- 1.Bandt, C., Pompe, B.: Permutation entropy – a natural complexity measure for time series. Phys. Rev. Lett. 88(17) (April 2002)Google Scholar
- 5.Cao, Y., Wen Tung, W., Gao, J.B., Protopopescu, V.A., Hively, L.M.: Detecting dynamical changes in time series using the permutation entropy. Phys. Rev. E 70(4) (October 2004)Google Scholar
- 7.Fawcelt, T.: Roc graphs: Notes and practical considerations for data mining researchers. In: Intelligent Enterprise Technologies Laboratory HP Laboratories Palo Alto, HPL-2003-4 (January 2003)Google Scholar
- 8.Gao, D., Hu, J., Buckley, T., White, K., Hass, C.: Shannon and Renyi entropy to classify effects of mild traumatic brain injury on postural sway. PLoS One 6(9) (2011)Google Scholar
- 11.Kahraman, S., Dutton, R.P., Hu, P., Stansbury, L., et al.: Heart rate and pulse pressure variability are associated with intractable intracranial hypertension after severe traumatic brain injury. Clinical investigation 22(4) (October 2010)Google Scholar
- 12.Kahraman, S., Hu, P., Stein, D., Stansbury, L., Dutton, R., Xiao, Y., Hess, J., Scalea, T.: Dynamic three-dimensional scoring of cerebral perfusion pressure and intracranial pressure provides a brain trauma index that predicts outcome in patients with severe traumatic brain injury. J. Trauma 70(3), 547–553 (2011)CrossRefGoogle Scholar
- 15.Lopes, F.M., de Oliveira, E.A., Cesar, J.R.M.: Inference of gene regulatory networks from time series by Tsallis entropy. BMC Systems Biology 5(61) (2011)Google Scholar
- 19.Platt, J.C.: Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods. In: Advances in Large Margin Classifiers, pp. 61–74. MIT Press (1999)Google Scholar
- 20.Provost, F., Domingos, P.: Well-trained pets: Improving probability estimation trees (2000)Google Scholar
- 23.Zanin, M.: Forbidden patterns in financial time series. Chaos 18(1), 013119 (2008)Google Scholar