Support Vector Machine with External Recurrences for Modeling Dynamic Cerebral Autoregulation

  • Max Chacón
  • Darwin Diaz
  • Luis Ríos
  • David Evans
  • Ronney Panerai
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4225)


Support Vector Machines (SVM) have been applied extensively to classification and regression problems, but there are few solutions proposed for problems involving time-series. To evaluate their potential, a problem of difficult solution in the field of biological signal modeling has been chosen, namely the characterization of the cerebral blood flow autoregulation system, by means of dynamic models of the pressure-flow relationship. The results show a superiority of the SVMs, with 5% better correlation than the neural network models and 18% better than linear systems. In addition, SVMs produce an index for measuring the quality of the autoregulation system which is more stable than indices obtained with other methods. This has a clear clinical advantage.


Support Vector Machine biological signals cerebral autoregulation 


  1. 1.
    Raicharoen, T., Lursinsap, C., Sanguanbhokai, P.: Application of critical support vector machine to time series prediction. In: Circuits and Systems. Proceedings of ISCAS 2003, vol. 5, pp. 741–744 (2003)Google Scholar
  2. 2.
    Wan-Zhao, C., Chang-Chun, Z., Wen-Xing, B., Jun-Hua, L.: Chaotic time series prediction using mean-field theory for support vector machine. Chinese Phys. 14, 922–929 (2005)CrossRefGoogle Scholar
  3. 3.
    Jankowski, S., Oreziak, A.: Learning system for computer-aided ECG analysis based on support vector machines. International Journal of Bioelectromagnetism 5, 175–176 (2003)Google Scholar
  4. 4.
    Acir, N., Guzelis, C.: Automatic spike detection in EEG by a two-stage procedure based on support vector machines. Comput. Biol. Med. 7, 561–575 (2004)CrossRefGoogle Scholar
  5. 5.
    Newell, D., Aaslid, R., Lam, A., Mayberg, T., Winn, R.: Comparison of flow and velocity during autoregulation testing in humans. Stroke 25, 793–797 (1994)Google Scholar
  6. 6.
    Panerai, R.: Assessment of cerebral pressure autoregulation in humans - a review of measurement methods. Physiological Measurement 19, 305–338 (1998)CrossRefGoogle Scholar
  7. 7.
    Panerai, R., Evans, D., Mahony, P., Deverson, S., Hayes, P.: Assessment of thigh cuff technique for measurement of dynamic cerebral autoregulation. Stroke 31, 476–480 (2000)Google Scholar
  8. 8.
    Panerai, R.B., Dawson, S.L., Eames, P.J., Potter, J.F.: Cerebral blood flow velocity response to induced and spontaneous sudden changes in arterial blood pressure. Am J Physiol. 280, H2162–H2174 (2001)Google Scholar
  9. 9.
    Tiecks, F., Lam, A., Aalid, R., Newell, D.: Comparison of static and dynamic cerebral Autoregulation measurements. Stroke 26, 1014–1019 (1995)Google Scholar
  10. 10.
    Panerai, R., Dawson, S., Potter, J.: Linear and nonlinear analysis of human dynamic cerebral autoregulation. Am J Physiol. 227, H1089–H1099 (1999)Google Scholar
  11. 11.
    Panerai, R., Chacón, M., Pereira, R., Evans, D.: Neural network modeling of dynamic Cerebral Autoregulation: assessment and comparison with established methods. Med. Eng & Phys. 26, 43–52 (2004)CrossRefGoogle Scholar
  12. 12.
    Mitsis, G., Zhang, G., Levine, B.D., Marmarelis, V.: Modeling of Nonlinear Physiological Systems with fast and Slow Dynamics. II. Application to cerebral Autoregulation. Ann. Biomedical Engineering 30, 555–565 (2002)Google Scholar
  13. 13.
    Schölkopf, B., Smola, A., Williamson, R.C., Bartlett, P.: New support vector algorithms. Neural Computation 12, 1207–1245 (2000)CrossRefGoogle Scholar
  14. 14.
    Vapnik, V.: The Nature of Statistical Learning Theory. Springer, New York (1995)zbMATHGoogle Scholar
  15. 15.
    Nerrand, O., Roussel-Ragot, P., Personnaz, L., Dreyfus, G.: Neural networks and non-linear adaptive filtering: unifying concepts and new algorithms. Neural Comput. 5, 165–199 (1993)CrossRefGoogle Scholar
  16. 16.
    Frohlich, H., Zell, A.: Efficient parameter selection for support vector machines in classification and regression via model-based global optimization. In: Neural Networks. IJCNN 2005. Proceedings. 2005 IEEE International Joint Conference, vol. 3, pp. 1431–1436 (2005)Google Scholar
  17. 17.
    Panerai, R.B., Eames, P.J., Potter, J.F.: Variability of time-domain indices of dynamic cerebral Autoregulation. Physiol. Meas. 24, 367–381 (2003)CrossRefGoogle Scholar
  18. 18.
    Chacón, M., Blanco, C., Panerai, R., Evans, D.: Nonlinear Modeling of Dynamic Cerebral Autoregulation Using Recurrent Neural Networks. In: Sanfeliu, A., Cortés, M.L. (eds.) CIARP 2005. LNCS, vol. 3773, pp. 205–213. Springer, Heidelberg (2005)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Max Chacón
    • 1
  • Darwin Diaz
    • 1
  • Luis Ríos
    • 1
  • David Evans
    • 2
  • Ronney Panerai
    • 2
  1. 1.Departamento de Ingeniería InformáticaUniversidad de Santiago de ChileCasillaChile
  2. 2.Medical Physics Group, Department of Cardiovascular SciencesUniversity of Leicester, Leicester Royal InfirmaryLeicesterUK

Personalised recommendations