Neural Networks for Biomedical Signals Classification Based on Empirical Mode Decomposition and Principal Component Analysis

  • Abdoul Dalibou Abdou
  • Ndeye Fatou NgomEmail author
  • Samba Sidibé
  • Oumar Niang
  • Abdoulaye Thioune
  • Cheikh H. T. C. Ndiaye
Conference paper
Part of the Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering book series (LNICST, volume 204)


The three main events presented in the electrocardiogram (ECG) signal of each heartbeat are: the P wave, the QRS complex and the T wave. Each event contains its own peak, making this important to analyze their morphology, amplitude and duration for cardiac abnormalities. In this study, we propose a system for biomedical signal analysis based on empirical mode decomposition. Mustispectral analysis is first performed to remove noise, detect QRS complex and compute the QRS wide. Then statistical features and QRS wide are after used as inputs of classifier based on neural network model. The proposed methodology is tested on real biomedical data and discussed.


Empirical mode decomposition Neural network ECG signal classification 


  1. 1.
    Al-Ashkar, T.: Medical multi signal signature recognition applied to cardiac diagnostic (2012)Google Scholar
  2. 2.
    Chang, P.C., Lin, J.J., Hsieh, J.C., Weng, J.: Myocardial infarcation classification with multi-lead ECG using hidden Markov models and Gaussian mixture models. Appl. Soft Comput. 12, 3165–3175 (2012)CrossRefGoogle Scholar
  3. 3.
    Fatin, A.E., Salim, N., Harris, A.R., Swee, T.T., Ahmed, T.: Arrhythmia recognition and classification using combined linear and nonlinear feature of ECG signals. Comput. Methods Programs Biomed. 127, 52–63 (2016)CrossRefGoogle Scholar
  4. 4.
    Jin, F., Sugavaneswaran, L., Krishnan, S., Chauhan, V.S.: Quantification of fragmented QRS complex using intrinsic time-scale decomposition. Biomed. Signal Process. Control 31, 513–523 (2017)CrossRefGoogle Scholar
  5. 5.
    Moody, G.B., Mark, R.G.: The impact of the MIT-BIH arrhythmia database. IEE Eng. Med. Biol. 20, 45–50 (2001)CrossRefGoogle Scholar
  6. 6.
    Niang, O., Thioune, A., Delechelle, E., Lemoine, J.: Spectral intrinsic decomposition method for adaptative signal representation. ISRN Sig. Process. 9, 3 (2012)zbMATHGoogle Scholar
  7. 7.
    Pal, S., Mitra, M.: Empirical mode decomposition based on ECG enhencement and QRS detection. Comput. Biol. Med. 42, 83–92 (2012)CrossRefGoogle Scholar
  8. 8.
    Pasa, L., Testolin, A., Sperdui, A.: Neural networks for sequential data: a pretraining approach based on hidden Markov models. Neurocomputing 169, 323–333 (2015)CrossRefGoogle Scholar
  9. 9.
    Polat, K., Gunes, S.: Detection of ECG arrythmia using a differential expert system approach based on principal component analysis and least square support vector machine. Appl. Math. Comput. 186, 898–906 (2007)MathSciNetzbMATHGoogle Scholar
  10. 10.
    Ripoll, V.J.R., Wojdel, A., Romero, E., Ramos, P., Brugada, J.: ECG assessment based on neural networks with pretraining. Appl. Soft Comput. 49, 399–406 (2016)CrossRefGoogle Scholar
  11. 11.
    Rodriguez, R., Mexicano, A., Bila, J., Cervantes, S., Ponce, R.: Feature extraction of electrocardiogram signals by aplying adaptative threshold and principal component analysis. J. Appl. Res. Technol. 13, 261–269 (2015)CrossRefGoogle Scholar
  12. 12.
    Slimane, Z.H., Nait Ali, A.: QRS complex detection using empirical mode decomposition. Digit. Signal Proc. 20, 1221–1228 (2010)CrossRefGoogle Scholar
  13. 13.
    Tjolleng, A., Jung, K., Hong, W., Lee, W., Lee, B., You, H., Son, J., Park, S.: Classification of a driver’s cognitive workload levels using artificial neural network on ECG signals. Appl. Ergon. 59, 326–332 (2017)CrossRefGoogle Scholar
  14. 14.
    Tu, J.V.: Advantages and disadvantages of using artificial neural networks versus logistic regression for predicting medical outcomes. J. Clin. Epidemiol. 45, 1225–1231 (1996)CrossRefGoogle Scholar
  15. 15.
    Zhang, H.: An improved QRS wave group detection algorithm and matlab implementation. Phys. Procedia 25, 1010–1016 (2012)CrossRefGoogle Scholar

Copyright information

© ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2018

Authors and Affiliations

  • Abdoul Dalibou Abdou
    • 2
  • Ndeye Fatou Ngom
    • 1
    Email author
  • Samba Sidibé
    • 1
  • Oumar Niang
    • 1
  • Abdoulaye Thioune
    • 3
  • Cheikh H. T. C. Ndiaye
    • 3
  1. 1.Laboratoire Traitement de l’Inforrmation et Systémes Intelligents (LTISI)Ecole Polytechnique de Thies (EPT)ThiesSenegal
  2. 2.Univerté de Thies, LTISI-EPTThiesSenegal
  3. 3.Univerité Cheikh Anta Diop de Dakar, LTISI-EPTDakarSenegal

Personalised recommendations