Advertisement

A New Wavelet-Based Neural Network for Classification of Epileptic-Related States using EEG

  • E. Juárez-Guerra
  • V. Alarcon-Aquino
  • P. Gómez-GilEmail author
  • J. M. Ramírez-Cortés
  • E. S. García-Treviño
Article

Abstract

In this paper, we present a novel neural network able to classify epileptic seizures using electroencephalogram (EEG) signals, called “Multidimensional Radial Wavelons Feed-Forward Wavelet Neural Network” (MRW-FFWNN). The network is part of a classification system, which distinguishes among three brain states related to epilepsy namely ictal, interictal and healthy. Efficient methods for pre-processing EEG’s, extracting features and getting the final class decisions were selected using a statistical three-fold cross-validation method, which assures the robustness of the system and its generalization ability. The following methods were systematically analyzed to find the most appropriate for this problem: 1) Infinite Impulse Response (IIR) and Finite Impulse Response (FIR) filters for noise reduction; 2) discrete Wavelet Transform (DWT) and Maximal Overlap Discrete Wavelet Transform (MODWT) for frequency decomposition of the EEG signals; 3) average correlation and maximum voting correlation for selecting a suitable mother wavelet for frequency decomposition; 4) Binary-tree and one-vs-one (OVO) decomposition strategies for primary binary classification; 5) voting and weighted-voting strategy aggregation strategies for the final classification. The integrated system was assessed using a three-fold cross validation, applied to a benchmark provided by the University of Bonn, getting an average accuracy of 93.33% when tested using sets Z, S and F and 95.0% when sets Z, S, F and O were used. The proposed network got competitive accuracy, compared with other state-of-the art classifiers, training in almost a half of the time than the ones with similar accuracy.

Keywords

Wavelet-based neural networks Epileptic seizure detection EEG analysis Machine learning classification Wavelet selection 

Notes

Acknowledgments

The first author gratefully acknowledges the financial support from the Universidad Autónoma de Tlaxcala and the Teacher Improvement Program (PROMEP) by scholarship No. UATLX-244. This research was partially supported by the National Council of Science and Technology in México (CONACYT), project grant No. CB-2010-155250.

References

  1. 1.
    Acharya, U.R., Sree, S.V., Swapna, G., Martis, R.J., Suri, J.S. (2013). Automated EEG analysis of epilepsy: A review. Knowledge-based Systems, 45, 147–165.  https://doi.org/10.1016/j.knosys.2013.02.014. http://www.sciencedirect.com/science/article/pii/S0950705113000798.Google Scholar
  2. 2.
    Alarcon-Aquino, V., & Barria, J. (2009). Change detection in time series using the maximal overlap discrete wavelet transforms. Latin American Applied Research: An International Journal, 39(2), 145–152.Google Scholar
  3. 3.
    Alarcon-Aquino, V., Ramírez-Cortés, J., Gómez-Gil, P., Starostenko, O., García-González, Y. (2014). Network intrusion detection using self-recurrent wavelet-neural-network with multidimensional radial wavelon. Information Technology and Control, 43(4), 347–358.Google Scholar
  4. 4.
    Andrzejak, R., Lehnertz, K. , Rieke, C., Mormann, F., David, P., Elger, C. (2001). Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: Dependence on recording region and brain state. Phy. Rev, 64(6), 061907–1–061907-8.Google Scholar
  5. 5.
    Anusha, K.S., Mathew, T.M., Subha, D.P. (2012). Classification of normal and epileptic EEG signal using time & frequency domain features through artificial neural network. In: 2012 international conference on advances in computing and communications (pp. 98–101).Google Scholar
  6. 6.
    Avinash, J., & Rashmi, D. (2014). Efficient method for optimizing artificial neural network using ‘quantum-based algorithm’. International Journal of Advanced Research in Computer Science and Software Engineering, 4(6), 692–700.Google Scholar
  7. 7.
    Birkemeier, W.P., Fontaine, A.B., Celesia, G.G., Ma, K.M. (1978). Pattern recognition techniques for the detection of epileptic transients in eeg. IEEE Transactions on Biomedical Engineering, BME-25(3), 213–217.  https://doi.org/10.1109/TBME.1978.326324.Google Scholar
  8. 8.
    Blinowska, K., & Durka, P. (2006). Electroencephalography (EEG). Wiley Encyclopedia of Biomedical Engineering, 1st edn. Hoboken: Wiley.Google Scholar
  9. 9.
    Bonn, U. (2016). Kinik für epiteptologie: EEG data download page, last accessed at march 25, 2016. Free resource, available in http://epileptologie-bonn.de/cms/front_content.php?idcat=495&idart=855.
  10. 10.
    Carrie, J. (1972). A hybrid computer system for detecting and quantifying spike and wave EEG patterns. Electroencephalography and Clinical Neurophysiology, 33(3), 339–341.  https://doi.org/10.1016/0013-4694(72)90164-2. http://www.sciencedirect.com/science/article/pii/0013469472901642.Google Scholar
  11. 11.
    Conlon, T., Ruskin, H., Crane, M. (2009). Seizure characterisation using frequency-dependent multivariate dynamics. Computers in Biology and Medicine, 39 (9), 760–767.  https://doi.org/10.1016/j.compbiomed.2009.06.003. http://www.sciencedirect.com/science/article/pii/S0010482509001103.Google Scholar
  12. 12.
    Cook, M.J., O’Brien, T.J., Berkovic, S.F., Murphy, M., Morokoff, A., Fabinyi, G., D’Souza, W., Yerra, R., Archer, J., Litewka, L., Hosking, S., Lightfoot, P., Ruedebusch, V., Sheffield, W.D., Snyder, D., Leyde, K., Himes, D. (2013). Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study. The Lancet Neurology, 12(6), 563–571.  https://doi.org/10.1016/S1474-4422(13)70075-9. http://www.sciencedirect.com/science/article/pii/S1474442213700759.Google Scholar
  13. 13.
    Das, A.B., & Bhuiyan, M.I.H. (2016). Discrimination and classification of focal and non-focal EEG signals using entropy-based features in the emd-dwt domain. Biomedical Signal Processing and Control, 29, 11–21.  https://doi.org/10.1016/j.bspc.2016.05.004. http://www.sciencedirect.com/science/article/pii/S1746809416300465.Google Scholar
  14. 14.
    Das, A.B., Bhuiyan, M.I.H., Alam, S.M.S. (2016). Classification of EEG signals using normal inverse gaussian parameters in the dual-tree complex wavelet transform domain for seizure detection. Signal, Image and Video Processing, 10(2), 259–266.  https://doi.org/10.1007/s11760-014-0736-2.Google Scholar
  15. 15.
    Daubechies, I. (2004). Ten Lectures on Wavelets, New York: Society for Industrial and Applied Mathematics, 8th edn. USA: CBMS-NSF Regional Conference Series in Applied Mathematics.Google Scholar
  16. 16.
    Dietterich, T. (1998). Approximate statistical tests for comparing supervised classification learning algorithms. Neural Computation, 10, 1895–1923.Google Scholar
  17. 17.
    Duque-Munoz, L., Espinosa-Oviedo, J., Castellanos-Domínguez, C. (2014). Identification and monitoring of brain activity based on stochastic relevance analysis of short-time EEG rhythms. BioMedical Engineering Online, 13(1), 123–142.Google Scholar
  18. 18.
    Everitt, B. (1977). The Analysis of Contingency Tables. New York: Chapman and Hall.zbMATHGoogle Scholar
  19. 19.
    Frost, J.D.J. (1985). Automatic recognition and characterization of epileptiform discharges in the human EEG. Journal of Clinical Neurophysiology pp. 231–250. https://journals.lww.com/clinicalneurophys/Fulltext/1985/07000/Automatic_Recognition_and_Characterization_of.3.aspx.
  20. 20.
    Galar, M., Fernández, A., Barrenechea, E., Bustince, H., Herrera, F. (2011). An overview of ensemble methods for binary classifiers in multi-class problems: Experimental study on one-vs-one and one-vs-all schemes. Pattern Recognition, Elsevier., 44(1), 1761–1776.Google Scholar
  21. 21.
    Gandhi, T., Panigrahi, B., Anand, S. (2011). A comparative study of wavelet families for EEG signal classification. Neurocomputing, 74(1), 3051–3057.Google Scholar
  22. 22.
    García-González, Y. (2011). Master Thesis: Modelos y algoritmos para Redes Neuronales Recurrentes Basadas en Wavelets Aplicados a la detección de Intrusos, 1st edn. México: Department of Computing, Electronics and Mechatronics., Universidad de las Américas Puebla.Google Scholar
  23. 23.
    Ghosh, D., Adeli, H., Dadmehr, N. (2007). Mixed-band wavelet-chaos-neural network methodology for epilepsy and epileptic seizure detection. IEEE Transactions on Biomedical Engineering, 54(9), 1545–1551.Google Scholar
  24. 24.
    Gómez-Gil, P., Juárez-Guerra, E., Alarcón-Aquino, V., Ramírez-Cortés, M., Rangel-Magdaleno, J. (2014). Identification of Epilepsy Seizures Using Multi-resolution Analysis and Artificial Neural Networks, (pp. 337–351). Cham: Springer International Publishing.  https://doi.org/10.1007/978-3-319-05170-3_23.Google Scholar
  25. 25.
    Hagan, M.T., Demuth, H.B., Beale, M.H., Jesus, O.D. (2010). Neural Network Design. 2 edn, ISBN-10: 0-9717321-1-6, USA.Google Scholar
  26. 26.
    Hariharan, M., Fook, C., Ilias, B. (2012). A comparative study of wavelets families for classification of wrist motions. Computers and Electrical Engineering, Elsevier, 38(1), 1798–1807.Google Scholar
  27. 27.
    Hassan, A.R., & Subasi, A. (2016). Automatic identification of epileptic seizures from EEG signals using linear programming boosting. Computer Methods and Programs in Biomedicine, 136, 65–77.Google Scholar
  28. 28.
    Hunter, D., Yu, H., Kolbusz, M., Wilamowski, B. (2012). Selection of proper neural network sizes and architectures-a comparative study. IEEE Transactions on Industrial Informatics, 8(2), 228–240.Google Scholar
  29. 29.
    Jahankhani, P., Kodogiannis, V., Revett, K. (2006). EEG Signal classification using wavelet feature extraction and neural networks. In: IEEE John Vincent Atanasoff 2006 international symposium on modern computing (JVA’06) (pp. 120–124).  https://doi.org/10.1109/JVA.2006.17.
  30. 30.
    Jimenez-Guarneros, M., & Gomez-Gil, P. (2017). Cross-subject classification of cognitive loads using a recurrent-residual deep network. In: 2017 IEEE Symposium Series on Computational Intelligence (SSCI), pp. 1–7.  https://doi.org/10.1109/SSCI.2017.8280897.
  31. 31.
    Juárez-Guerra, E. (2016). Doctoral Thesis: Epileptic Seizures Classification in EEG using wavelet-based Neural Networks, 1st edn. México: Department of Computing, Electronics and Mechatronics., Universidad de las Américas Puebla.Google Scholar
  32. 32.
    Juárez-Guerra, E., Alarcon-Aquino, V., Gómez-Gil, P. (2015). Epilepsy seizure detection in EEG signals using wavelet transforms and neural networks. New Trends in Networking, Computing, E-learning, Systems Sciences, and Engineering, Lecture Notes in Electrical Engineering. Eds: K. elleithy, T, Sobh., 312(1), 261–269.Google Scholar
  33. 33.
    Juárez-Guerra, E., Gómez-Gil, P., Alarcon-Aquino, V. (2013). Biomedical signal processing using wavelet-based neural networks. Special Issue: Advances in Pattern Recognition, Research in Computing Science, 61 (1), 23–32.Google Scholar
  34. 34.
    Kelley, M., Jacobs, M., Lowenstein, D. (2009). The ninds epilepsy research benchmarks. Epilepsia, 50 (3), 579–582.  https://doi.org/10.1111/j.1528-1167.2008.01813.x.Google Scholar
  35. 35.
    Khan, M.M., Mendes, A., Zhang, P., Chalup, S.K. (2017). Evolving multi-dimensional wavelet neural networks for classification using cartesian genetic programming. Neurocomputing, 247, 39–58.  https://doi.org/10.1016/j.neucom.2017.03.048. http://www.sciencedirect.com/science/article/pii/S0925231217305635.Google Scholar
  36. 36.
    Kocadagli, O., & Langari, R. (2017). Classification of EEG signals for epileptic seizures using hybrid artificial neural networks based wavelet transforms and fuzzy relations. Expert Systems with Applications, 88, 419–434.  https://doi.org/10.1016/j.eswa.2017.07.020. http://www.sciencedirect.com/science/article/pii/S0957417417304967.Google Scholar
  37. 37.
    Kohavi, R. (1995). A study of cross-validation and bootstrap for accuracy estimation and model selection. Proceedings of the International Joint Conference on Artificial Intelligence, 14(1), 1137–1145.Google Scholar
  38. 38.
    Kumari, A., & Bisht, M. (2013). Optimal wavelet filter maximizes the cross correlation coefficient with an ECG signal. International Journal of Innovative Technology and Research, 1(2), 191–193.Google Scholar
  39. 39.
    Lahiri, S., & Ghanta, K. (2009). Artificial neural network model with the parameter tuning assisted by a differential evolution technique: The study of the hold up of the slurry flow in a pipeline. Chemical Industry & Chemical Engineering Quarterly, 15(2), 103–117.Google Scholar
  40. 40.
    Li, M., Chen, W., Zhang, T. (2017). Application of modwt and log-normal distribution model for automatic epilepsy identification. Biocybernetics and Biomedical Engineering, 37(4), 679–689.  https://doi.org/10.1016/j.bbe.2017.08.003. http://www.sciencedirect.com/science/article/pii/S0208521617301754.Google Scholar
  41. 41.
    Li, M., Chen, W., Zhang, T. (2017). Classification of epilepsy EEG signals using dwt-based envelope analysis and neural network ensemble. Biomedical Signal Processing and Control, 31, 357–365.  https://doi.org/10.1016/j.bspc.2016.09.008. http://www.sciencedirect.com/science/article/pii/S1746809416301343.Google Scholar
  42. 42.
    Mandic, D., Rehman, N., Wu, Z., Huang, N. (2013). Empirical mode decomposition-based time-frequency analysis of multivariate signals. IEEE Signal Processing Magazine, 30(6), 74–86.Google Scholar
  43. 43.
    Martis, R., Acharya, U., Tan, J., Petznick, A., Yanti, R., Chua, C. (2012). Application of empirical mode decomposition (EMD) for automated detection of epilepsy using EEG signals. International Journal of Neural Systems, 22(6), 1250027–1–1250027-16.Google Scholar
  44. 44.
    Mathworks, T. (2013). Matlab Documentation, 7th edn. Natick: The Mathworks, Inc.Google Scholar
  45. 45.
    Mirzaei, A., Ayatollahi, A., Vavadi, H. (2011). Statistical analysis of epileptic activities based on histogram and wavelet-spectral entropy. Biomedical Science and Engineering, 4(1), 207–213.Google Scholar
  46. 46.
    Mitchell, T. (1997). Machine Learning, 1st edn. New York: McGraw-Hill Science.zbMATHGoogle Scholar
  47. 47.
    Peker, M., Sen, B., Delen, D. (2016). A novel method for automated diagnosis of epilepsy using complex-valued classifiers. IEEE Journal of Biomedical and Health Informatics, 20(1), 108–118.Google Scholar
  48. 48.
    Percival, D., & Walden, A. (2000). Wavelet Methods for Time Series Analysis, 1st edn. Cambridge: Cambridge University Press.zbMATHGoogle Scholar
  49. 49.
    Proakis, J., & Manolakis, D. (2007). Digital Signal Processing, 4th edn. USA: Prentice Hall.Google Scholar
  50. 50.
    Qazi, K.I., Lam, H., Xiao, B., Ouyang, G., Yin, X. (2016). Classification of epilepsy using computational intelligence techniques. CAAI Transactions on Intelligence Technology, 1(2), 137–149.Google Scholar
  51. 51.
    Ramgopal, S., Thome-Souza, S., Jackson, M., Kadish, N.E., Fernández, I. S., Klehm, J., Bosl, W., Reinsberger, C., Schachter, S., Loddenkemper, T. (2014). Seizure detection, seizure prediction, and closed-loop warning systems in epilepsy. Epilepsy & Behavior, 37, 291–307.  https://doi.org/10.1016/j.yebeh.2014.06.023. https://www.sciencedirect.com/science/article/pii/S1525505014002297.Google Scholar
  52. 52.
    Ravish, D., & Devi, S. (2012). Automated seizure detection and spectral analysis of EEG seizure time series. European Journal of Scientific Research, 68(1), 72–82.Google Scholar
  53. 53.
    Samarasinghe, S. (2007). Neural Networks for Applied Sciences and Engineering From Fundamentals to Complex Pattern Recognition, 1st edn. USA: Auerbach Publications.zbMATHGoogle Scholar
  54. 54.
    Santhanam, T., & Padmavathi, M. (2015). Application of k-means and genetic algorithms for dimension reduction by integrating SVM for diabetes diagnosis. Computer Science Elsevier, 47(1), 76–83.Google Scholar
  55. 55.
    Saraswathy, J., Hariharan, M., Nadarajaw, T., Khairunizam, W., Yaacob, S. (2014). Optimal selection of mother wavelet for accurate infant cry classification. Technical paper Australasian College of Physical Scientists and Engineers in Medicine, 37(2), 439–456.Google Scholar
  56. 56.
    Satapathy, S.K., Dehuri, S., Jagadev, A.K. (2017). EEG Signal classification using pso trained rbf neural network for epilepsy identification. Informatics in Medicine Unlocked, 6, 1–11.  https://doi.org/10.1016/j.imu.2016.12.001. http://www.sciencedirect.com/science/article/pii/S2352914816300387.Google Scholar
  57. 57.
    Shaik, J., & Srinivasa, K. (2012). Epileptic seizures classification from EEG signals using neural networks. International Conference on Information and Network Technology, 37(1), 269–273.Google Scholar
  58. 58.
    Sheng-Fu, L., HsuChuan, W., Wan-Lin, C. (2010). Combination of EEG complexity and spectral analysis for epilepsy diagnosis and seizure detection. EURASIP Journal on Advances in Signal Processing, 2010(62), 853434–1–853434-15.Google Scholar
  59. 59.
    Singh, B., & Tiwari, A. (2006). Optimal selection of wavelet basis function applied to ECG signal denoising. Digit Signal Process, 16(3), 275–287.Google Scholar
  60. 60.
    Song, Y., Crowcroft, J., Zhang, J. (2012). Automatic epileptic seizure detection in EEGs based on optimized sample entropy and extreme learning machine. Journal of Neuroscience Methods, 210(2), 132–146.  https://doi.org/10.1016/j.jneumeth.2012.07.003. http://www.sciencedirect.com/science/article/pii/S0165027012002671.Google Scholar
  61. 61.
    Sriram, R., Thome-Souza, S., Jackson, M., Kadish, N., Sanchez-Fernandez, I., Klehma, J., Bosl, W., Reinsberger, C., Schachter, S., Loddenkemper, T. (2014). Seizure detection, seizure prediction, and closed-loop warning systems in epilepsy. Epilepsy and Behavior, 37(1), 291–307.Google Scholar
  62. 62.
    Subasi, A., & Ercelebi, E. (2005). Classification of EEG signals using neural network and logistic regression. Computer Methods and Programs in Biomedicine, 1(78), 87–99.zbMATHGoogle Scholar
  63. 63.
    Sung, J., Jin, B., Yoon, H. (2005). Direct adaptive control using self recurrent wavelet neural network via adaptive learning rates for stable path tracking of mobile robots. Proceedings of the 2005 American Control Conference, Portland, 1(1), 288–293.Google Scholar
  64. 64.
    Sung, J., Jin, B., Yoon, H. (2007). Indirect adaptive control of nonlinear dynamic systems using self recurrent wavelet neural network via adaptive learning rates. Proceedings of Information Sciences, 177(1), 3074–3098.MathSciNetzbMATHGoogle Scholar
  65. 65.
    Sunhaya, S., & Manimegalai, S. (2012). Detection of epilepsy disorder in EEG signal. International Journal of Emerging and Development, 2(2), 473–479.Google Scholar
  66. 66.
    Tzallas, A.T., Tsipouras, M., Fotiadis, D. (2009). Epileptic seizure detection in EEGs using time-frequency analysis. IEEE Transactions on Information Technology in Biomedicine, 13(5), 703–710.Google Scholar
  67. 67.
    Tzallas, A.T., Tsipouras, M.G., Tsalikakis, D.G., Karvounis, E.C., Astrakas, L., Konitsiotis, S., Tzaphlidou, M. (2012). Automated epileptic seizure detection methods: a review study intech open science.  https://doi.org/10.5772/31597. https://mts.intechopen.com/books/epilepsy-histological-electroencephalographic-and-psychological-aspects/automated-epileptic-seizure-detection-methods-a-review-study.
  68. 68.
    Wang, Y., Zhou, W., Yuan, Q. (2013). Comparison of ictal and interictal EEG signals using fractal features. International Journal of Neural Systems, 23(6), 1350028.Google Scholar
  69. 69.
    Xie, S., & Krishnan, S. (2013). Wavelet-based sparse functional linear model with applications to EEGs seizure detection and epilepsy diagnosis. Medical &, Biological Engineering & Computing, 51(1), 49–60.  https://doi.org/10.1007/s11517-012-0967-8.Google Scholar
  70. 70.
    Zhang, Q. (1993). Wavelet network: The radial structure and an efficient initialization procedure. Technical report, Linkoping University 1423(1), 1–17.Google Scholar
  71. 71.
    Zhang, Q., & Benveniste, A. (1992). Wavelet networks. IEEE Transactions on Neural Networks, 3(6), 889–898.Google Scholar
  72. 72.
    Zhang, T., Chen, W., Li, M. (2018). Fuzzy distribution entropy and its application in automated seizure detection technique. Biomedical Signal Processing and Control, 39, 360–377.  https://doi.org/10.1016/j.bspc.2017.08.013, http://www.sciencedirect.com/science/article/pii/S1746809417301763.Google Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Universidad Autónoma de TlaxcalaTlaxcalaMexico
  2. 2.Universidad de las AméricasPueblaMexico
  3. 3.Instituto Nacional de Astrofísica Óptica y ElectrónicaPueblaMexico
  4. 4.Instituto de Investigaciones en Ecosistemas y SustentabilidadUNAMMoreliaMéxico

Personalised recommendations