Artificial Neural Network Models for Timely Assessment of Trauma Complication Risk

  • R. P. Marble
  • J. C. Healy
Part of the Studies in Fuzziness and Soft Computing book series (STUDFUZZ, volume 96)


This chapter espouses the deployment of neural network-based diagnostic aids for evaluation of morbidity risks in the prehospital, acute care, and rehabilitation circumstances evinced by traumatic injury. The potential effectiveness of such systems is addressed from several points of view. First, the ability of the underlying connectionist models to identify complex, highly nonlinear, and sometimes even counterintuitive patterns in trauma data is discussed. Prior work in the area is reviewed and the approach is illustrated with an application that succeeds in identifying coagulopathy outcomes in victims of blunt injury trauma. Second, the feasibility of the universal applicability of neural models in actual trauma situations is argued. Their ability to use standardized, widely available data and their capacity for reflecting local differences and changing conditions is exposed. Finally, the potential enhancements for such models are explored in the contexts of clinical decision support systems.


Neural Network Artificial Neural Network Hide Layer Neural Network Model Artificial Neural Network Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    McCulloch, W.S. and Pitts, W.H. (1943), “A logical calculus of the ideas immanent in nervous activity,” Bull. Math. Biophys., vol. 5, pp. 115–133.CrossRefGoogle Scholar
  2. [2]
    Hebb, D.O. (1949), The Organization of Behavior, John Wiley and Sons, New York.Google Scholar
  3. [3]
    Widrow, B. and Hoff, M.E. (1960), “Adaptive switching circuits,” IRE Wescon Conv. Rec., pp. 96–104.Google Scholar
  4. [4]
    Rosenblatt, F. (1961), Principles of Neurodynamics: Perceptrons and the Theory of Brain Mechanisms, Spartan Press, Washington, D.C.Google Scholar
  5. [5]
    Minsky, M. and Papert, S. (1969), Perceptrons: An Introduction to Computational Geometry, MIT Press, Cambridge, Mass.Google Scholar
  6. [6]
    Rumelhart, D.E. and McClelland, J.L. (1986), Parallel Distributed Processing, Vol. I and II, MIT Press, Cambridge, Mass.Google Scholar
  7. [8]
    Hagan, M.T., Demuth, H.B. and Beale, M. (1996), Neural Network Design, PWS Publishing Company, Boston.Google Scholar
  8. [9]
    Widrow, B., Rumelhart, D.E. and Lehr, M.A. (1994), “Neural networks: Applications in industry, business, and science,” CACM, vol. 37 (3), pp. 93–105.CrossRefGoogle Scholar
  9. [10]
    Ripley, B.D. (1993), “Statistical aspects of neural networks,” Networks and Chaos — Statistical and Probabilistic Aspects (Barndorff-Nielsen, O.E., Jensen, J.L. and Kendall, W.S., eds.), Chapman and Hall, London, pp. 40–123.Google Scholar
  10. [11]
    Cheng, B. and Titterington, D.M. (1994), “Neural networks: A review from a statistical perspective,” Statist. Sci., vol. 9 (1), pp. 254.Google Scholar
  11. [12]
    Warner, B. and Misra, M. (1996), “Understanding neural networks as statistical tools,” The American Statistician, vol. 50, pp. 284293.Google Scholar
  12. [13]
    Wang, S. (1998), “An insight into the standard backpropagation neural network model for regression analysis,” Omega, vol. 26, pp. 133–140.CrossRefGoogle Scholar
  13. [14]
    Hwang, J.T. and Ding, A.A. (1997), “Prediction intervals for artificial neural networks, ” J. Am. Statist. Assoc., vol. 92, pp. 748757.Google Scholar
  14. [15]
    Yoon, Y., Swales, G. and Margavio, T.M. (1993), “A comparison of discriminant analysis versus artificial neural networks,” J. Opl. Res. Soc., vol. 44, pp. 51–60.Google Scholar
  15. [16]
    Markham, I.S. and Rakes, T.R. (1998), “The effect of sample size and variability of data on the comparative performance of artificial neural networks and regression,” Computers Ops. Res., vol. 25, pp. 251–263.CrossRefGoogle Scholar
  16. [17]
    Cooper, J.C.B. (1999), “Artificial neural networks versus multivariate statistics: An application from economics,” J. Appl. Statist.,vol. 26, pp. 909–921.Google Scholar
  17. [18]
    Marble, R.P. and Maier, F.H. (1999), “Distribution-independent confidence intervals for non-linear models: An application to manufacturing performance factors,” Proceedings of the 29 th Annual Meeting of the Decision Sciences Institute, pp. 1232–1234.Google Scholar
  18. [19]
    Denton, J.W. (1995), “How good are neural networks for causal forecasting?” J. Bus. Forecasting, Summer, pp. 17–20.Google Scholar
  19. [20]
    Cybenko, G. (1989), “Approximation by superpositions of a sigmoidal function,” Mathematics of Control, Signals, and Systems, vol. 2, pp. 303–314.Google Scholar
  20. [21]
    Baxt, W.G. (1994), “Complexity, chaos and human physiology: The justification for non-linear neural computational analysis,” Cancer Lett., vol. 77, pp. 85–93.PubMedCrossRefGoogle Scholar
  21. [22]
    Baxt, W.G. (1991), “Use of an artificial neural network for the diagnosis of myocardial infarction,” Ann. Intern. Med., vol. 115, pp. 843–48.PubMedCrossRefGoogle Scholar
  22. [23]
    Downs, J., Harrison, R.F., Kennedy, R.L. and Cross,S.S. (1996), “Application of the fuzzy ARTMAP neural network model to medical pattern classification tasks,” Artificial Intelligence in Med., vol. 8, pp. 403–428.CrossRefGoogle Scholar
  23. [24]
    Mango, L.J. (1994), “Computer-assisted cervical cancer screening using neural networks,” Cancer Lett., vol. 77, pp. 155–162.PubMedCrossRefGoogle Scholar
  24. [25]
    Wilding, P., Morgan, M.A., Grygotis, A.E. et al. (1994), “Application of backpropagation neural networks to diagnosis of breast and ovarian cancer,” Cancer Lett., vol. 77, pp. 145–153.PubMedCrossRefGoogle Scholar
  25. [26]
    Lo, J.Y. et al. (1997), “Predicting breast cancer invasion with artificial neural networks on the basis of mammographic features,” Radiology, vol. 203, pp. 159–63.PubMedGoogle Scholar
  26. [27]
    Tourassi, G.D., Floyd, C.E., Sostman, H.D. and Coleman, R.E. (1993), “Acute pulmonary embolism: Artificial neural network approach for diagnosis,” Radiology, vol. 189, pp. 555–558.PubMedGoogle Scholar
  27. [28]
    Heden, B., Edenbrandt, L., Haisty, W.K. and Pahlm, O. (1994), “Artificial neural networks for the electrocardiographic diagnosis of healed myocardial infarction,” Am. J. Cardio., vol. 74, pp. 5–8.CrossRefGoogle Scholar
  28. [29]
    Kennedy, R.L. et al. (1997), “An artificial neural network system for diagnosis of acute myocardial infarction (AMI) in the accident and emergency department: Evaluation and comparison with serum myoglobin measurements,” Comput. Methods Programs Biomed., vol. 52, pp. 93–103.PubMedCrossRefGoogle Scholar
  29. [30]
    Orr, R.K. (1997), “Use of a probabilistic neural network to estimate the risk of mortality after cardiac surgery,” Medical Decision Making, vol. 17, pp. 178–185.PubMedCrossRefGoogle Scholar
  30. [31]
    Tian, J., Juhola, M. and Grönfors, T. (1997), “Latency estimation of auditory brainstem response by neural networks,” Artificial Intelligence in Medicine, vol. 10, pp. 115–128.PubMedCrossRefGoogle Scholar
  31. [32]
    Koprinska, I., Pfurtscheller, G. and Flotzinger, D. (1996), “Sleep classification in infants by decision tree-based neural networks,” Artificial Intelligence in Medicine, vol. 8, pp. 387–401.PubMedCrossRefGoogle Scholar
  32. [33]
    Henson, D.B., Spenceley, S.E. and Bull, D.R. (1997), “Artificial neural network analysis of noisy visual field data in glaucoma,” Artificial Intelligence in Medicine, vol. 10, pp. 99–113.PubMedCrossRefGoogle Scholar
  33. [34]
    Hosseini-Nezhad, S.M. et al. (1995), “A neural network approach for the determination of interhospital transport mode,” Computers and Biomed. Res., vol. 28, pp. 319–334.CrossRefGoogle Scholar
  34. [35]
    McGonigal, M.D., Cole, J., Schwab, C.W. et al. (1993), “A new approach to probability of survival scoring for trauma quality assurance,” J. Trauma, vol. 34, pp. 863–870.PubMedCrossRefGoogle Scholar
  35. [36]
    Boyd, C.R., Tolson, M.A. and Copes, W.S. (1987), “Evaluating trauma care: The TRISS method,” J. Trauma, vol. 27, pp. 370–378.PubMedCrossRefGoogle Scholar
  36. [37]
    Champion, H.R., Copes, W.S., Sacco, W.J. et al. (1990), “A new characterization of injury severity,” J. Trauma, vol. 30, pp. 539546.Google Scholar
  37. [38]
    Dombi, G.W. et al. (1995), “Prediction of rib fracture injury outcome by an artificial neural network,” J. Trauma, vol. 39, pp. 915–921.PubMedCrossRefGoogle Scholar
  38. [39]
    Izenberg, S.D., Williams, M.D. and Luterman, A. (1997), “Prediction of trauma mortality using a neural network,” Am. Surg., vol. 63, pp. 275–81.PubMedGoogle Scholar
  39. [40]
    Lim, C.P., Harrison, R.F. and Kennedy, R.L. (1997), “Application of autonomous neural network systems to medical pattern classification tasks,” Artificial Intelligence in Med., vol. 11, pp. 215–239.CrossRefGoogle Scholar
  40. [41]
    Selker, H.P. et al. (1995), “A comparison of performance of mathematical predictive methods for medical diagnosis: Identifying acute cardiac ischemia among emergency department patients,” J. Investig. Med., vol. 43, pp. 468–476.PubMedGoogle Scholar
  41. [42]
    Rutledge, R. (1995), “Injury severity and probability of survival assessment in trauma patients using a predictive hierarchical network model derived from ICD-9 codes,” J. Trauma, vol. 38, pp. 590–601.PubMedCrossRefGoogle Scholar
  42. [43]
    Champion, H.R., Copes, W.S., Sacco, W.J. et al. (1990), “The major trauma outcome study: Establishing national norms for trauma care,” J. Trauma, vol. 30, pp. 1356–1365.PubMedCrossRefGoogle Scholar
  43. [44]
    Marble, R.P. and Healy, J.C. (1999), “A neural network approach to the diagnosis of morbidity outcomes in trauma care,” Artificial Intelligence in Medicine, vol. 15, pp. 299–307.PubMedCrossRefGoogle Scholar
  44. [45]
    Rice, C.L. and Rutledge, R. (1993), “Trauma registry, in: American College of Surgeons - Committee on Trauma,” Resources for Optimal Care of the Injured Patient, pp. 97–101.Google Scholar
  45. [46]
    Hoyt, D.B., Hollingsworth-Fridlund, P., Fortlage, D. et al. (1992), “An evaluation of provider-related and disease-related morbidity in a level I university trauma service: Directions for quality improvement,” J. Trauma, vol. 33, pp. 586–601.PubMedCrossRefGoogle Scholar
  46. [47]
    Shapiro, M.J., Cole, K.E., Keegan, M., Prasad, C.N. and Thompson, R.J. (1994), “National survey of state trauma registries- 1992,” J. Trauma, vol. 37, pp. 835–842.PubMedCrossRefGoogle Scholar
  47. [48]
    Lett, R.R., Hanley, J.A. and Smith, J.S. (1995), “The comparison of injury severity instrument performance using likelihood ratio and ROC curve analyses,” J. Trauma, vol. 38, pp. 142–148.PubMedCrossRefGoogle Scholar
  48. [49]
    Skapura, D.M. (1995), Building Neural Networks, Addison-Wesley, Reading, Mass.Google Scholar
  49. [50]
    Hunter, A., Kennedy, L. et al. (2000), “Application of neural networks and sensitivity analysis to improved prediction of trauma survival,” Comput. Methods Programs Biomed., vol. 62, pp. 1119.CrossRefGoogle Scholar
  50. [51]
    Karnin, E.D. (1990), “A simple procedure for pruning back-propagation trained neural networks,” IEEE Transactions on Neural Networks, vol. 1, pp. 239–242.PubMedCrossRefGoogle Scholar
  51. [52]
    Chung, F.L. and Lee, L. (1992), “A node pruning algorithm for backpropagation networks,” International Journal of Neural Systems, vol. 3, pp. 301–314.CrossRefGoogle Scholar
  52. [53]
    Hagiwara, M. (1994), “A simple and effective method for removal of hidden units and weights,” Neurocomputing, vol. 6, pp. 207218.Google Scholar
  53. [54]
    Setiono, R. (1997), “A penalty-function approach for pruning feedforward networks,” Neural Computation, vol. 9, pp. 185–204.PubMedCrossRefGoogle Scholar
  54. [55]
    Towell, G.G. and Shavlik, J.W. (1993), “Extracting refined rules from knowledge-based neural networks,” Machine Learning, vol. 13, pp. 71–101.Google Scholar
  55. [56]
    Setiono, R., Thong, J.Y.L. and Yap, C. (1998), “Symbolic rule extraction from neural networks: An application to identifying organizations adopting IT,” Information and Management, vol. 34, pp. 91–101.CrossRefGoogle Scholar
  56. [57]
    Setiono, R. (1996) Extracting rules from pruned neural networks for breast cancer diagnosis,“ Artificial Intelligence in Medicine, vol. 8, pp. 37–51.PubMedCrossRefGoogle Scholar
  57. [58]
    Barret, J.P., Desai, M.H. and Herndon, D.N. (1999), “Massive transfusion of reconstituted whole blood is well tolerated in pediatric burn surgery,” J. Trauma, vol. 47, pp. 526–528.PubMedCrossRefGoogle Scholar
  58. [59]
    Fakhry, S.M., Messick, W.J. and Sheldon, G.F. (1996), “Metabolic effects of massive transfusion” in Principles of Transfusion Medicine, ( Rossi, E.C., Simon, T.L., Moss, G.S. and Gould, S.A. eds.), pp. 615–25, Williams and Wilkins, Baltimore.Google Scholar
  59. [60]
    Jones, J.M. (1995), “An approach to the analysis of trauma data having a response variable of death or survival,” J. Trauma, vol. 38, pp. 123–128.PubMedCrossRefGoogle Scholar
  60. [61]
    Lang, E.W. et al. (1997), “Outcome after severe head injury: an analysis of prediction based upon comparison of neural network versus logistic regression analysis,” Neurol. Res., vol. 19, pp. 274280.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • R. P. Marble
  • J. C. Healy

There are no affiliations available

Personalised recommendations