Advertisement

Annals of Biomedical Engineering

, Volume 38, Issue 1, pp 118–137 | Cite as

Selection of Dynamic Features Based on Time–Frequency Representations for Heart Murmur Detection from Phonocardiographic Signals

  • A. F. Quiceno-ManriqueEmail author
  • J. I. Godino-Llorente
  • M. Blanco-Velasco
  • G. Castellanos-Dominguez
Article

Abstract

This work discusses a method for the selection of dynamic features, based on the calculation of the spectral power through time applied to the detection of systolic murmurs from phonocardiographic recordings. To investigate the dynamic properties of the spectral power during murmurs, several quadratic energy distributions have been studied, namely Wigner-Ville, Choi-Williams, smoothed pseudo Wigner-Ville, exponential, and hyperbolic T-distribution. The classification performance has been compared with that using a Short Time Fourier Transform and Continuous Wavelet Transform representations. Furthermore, this work discusses a variety of nonparametric techniques to estimate the spectral power contours as dynamic features that characterize the heart sounds (HS): instantaneous energy, eigenvectors, instantaneous frequency, equivalent bandwidth, subband spectral centroids, and Mel cepstral coefficients. In this way, the aforementioned time–frequency representations and their dynamic features were evaluated by means of their ability to detect the presence of murmurs using a simple k-Nearest Neighbors classifier. Moreover, the relevancies of the proposed dynamic features have been evaluated using a time-varying principal component analysis. The work presented is carried out using a database containing 22 phonocardiographic recordings (16 normal and 6 records with murmurs), segmented to extract 402 representative individual beats (201 per class). The results suggest that the smoothing given by the quadratic energy distribution significantly improves the classification performance for the detection of murmurs in HS. Moreover, it is shown that the power dynamic features which give the best overall classification performance are the MFCC contours. As a result, the proposed method can be implemented as a simple diagnostic tool for primary health-care purposes with high accuracy (up to 98%) discriminating between normal and pathologic beats.

Keywords

Dynamic features Feature selection Heart sounds Time–frequency representations Automatic detection of murmurs 

Abbreviations

AUC

Area under ROC curve

CWD

Choi-Williams Distribution

CWT

Continuous wavelet transform

DTW

Dynamic time warping

EBW

Equivalent bandwidth

ECG

Electrocardiographic signal

ETD

Exponential T-distribution

HS

Heart sounds

HTD

Hyperbolic T-distribution

IF

Instantaneous frequency

MFCC

Mel Cepstral frequency coefficients

PCA

Principal component analysis

PCG

Phonocardiographic signal

ROC

Receiver operating characteristic curve

S1

First heart sound

S2

Second heart sound

SE

Standard error

SPWVD

Smoothed pseudo Wigner-Ville Distribution

STFT

Short Time Fourier Transform

t–f

Time–frequency

TFD

Time–frequency distribution

TFR

Time–frequency representation

WT

Wavelet transform

WVD

Wigner-Ville Distribution

ZPLN

Zero padding length normalization

Notes

Acknowledgments

The authors would like to acknowledge Dr. Ana Maria Matijasevič and Dr. Guillermo Agudelo, who are working with Hospital Infantil de Manizales: Rafael Henao Toro, for organizing the acquisition of the PCG data. This research was carried out under grants: “Jóvenes Investigadores”, funded by COLCIENCIAS; Centro de Investigación e Innovación de Excelencia ARTICA (Alianza Regional de TIC Aplicadas); and TEC2006-12887-C02 from the Ministry of Science and Technology of Spain. Moreover, the authors would like to thank the referees for their time, comments and dedication reviewing the manuscript. The authors acknowledge that their comments have clearly improved the manuscript.

References

  1. 1.
    Ahlstrom, C., P. Hult, P. Rask, J. E. Karlsson, E. Nylander, U. Dahlstrom, and P. Ask. Feature extraction for systolic heart murmur classification. Ann. Biomed. Eng. 34:1666–1677, 2006.CrossRefPubMedGoogle Scholar
  2. 2.
    Akay, M. Wavelet applications in medicine. IEEE Press 34(5):50–56, 1997.Google Scholar
  3. 3.
    Atlas, L., and J. Fang. Quadratic detectors for general nonlinear analysis of speech. In: Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing, 1992, pp. 9–12.Google Scholar
  4. 4.
    Boashash, B. Estimating and interpreting the instantaneous frequency of a signal. I. Fundamentals. IEEE J. Proc. 80:520–538, 1992.CrossRefGoogle Scholar
  5. 5.
    Cerutti, S., A. M. Bianchi, and L. T. Mainardi. Advanced spectral methods for detecting dynamic behaviour. Autonomic Neurosci. 90:3–12, 2001.CrossRefGoogle Scholar
  6. 6.
    Chakroborty, S., A. Roy, and G. Saha. Fusion of a complementary feature set with MFCC for improved closed set text-independent speaker identification. In: Proc. IEEE International Conference on Industrial Technology, 2006, pp. 387–390.Google Scholar
  7. 7.
    Chauhan, S., P. Wang, C. Sing Lim, and V. Anantharaman. A computer-aided MFCC-based HMM system for automatic auscultation. Comput. Biol. Med. 38:221–233, 2008.CrossRefPubMedGoogle Scholar
  8. 8.
    Cohen, L. The uncertainty principle in signal analysis. In: Proc. of the IEEE-SP International Symposium on Time–Frequency and Time-Scale Analysis, 1994, pp. 182–185.Google Scholar
  9. 9.
    Coifman, R. R., and M. V. Wickerhauser. Entropy-based algorithms for best basis selection. IEEE Trans. Inform. Theory 38:713–718, 1992.CrossRefGoogle Scholar
  10. 10.
    Daliman, S., and A. Z. Sháameri. Comparison of time–frequency analysis performance on hearts and murmurs. In: Proc. Student Conference on Research and Development, edited by A. Z. Sháameri, 2003, pp. 1–4.Google Scholar
  11. 11.
    Daliman, S., and A. Z. Sháameri. Time–frequency analysis of heart sounds and murmurs. In: Proc. of the 4th International Conference on Multimedia Information, Communications and Signal Processing, edited by A. Z. Sháameri, 2003, pp. 840–843.Google Scholar
  12. 12.
    Daliman, S., and A. Z. Sháameri. Time–frequency analysis of heart sounds using windowed and smooth windowed Wigner–Ville distribution. In: Proc. 7th International Symposium on Signal Processing and Its Applications, 2nd edn, edited by A. Z. Sháameri, 2003, pp. 625–626.Google Scholar
  13. 13.
    Daza-Santacoloma, G., J. D. Arias-Londono, J. I. Godino-Llorente, N. Saenz-Lechon, V. Osma-Ruiz, and C. G. Castellanos-Dominguez. Dynamic feature extraction: an application to voice pathology detection. Intell. Automat. Soft Comput. 15(4):665–680, 2009.Google Scholar
  14. 14.
    Debbal, S. M., and F. Bereksi-Reguig. Time–frequency analysis of the first and the second heartbeat sounds. Appl. Math. Comput. 184:1041–1052, 2007.CrossRefGoogle Scholar
  15. 15.
    Debbal, S. M., and F. Bereksi-Reguig. Analysis of the second cardiac sound using the fast Fourier and the continuous wavelet transforms. Internet J. Med. Technol. 3(1), 2006.Google Scholar
  16. 16.
    Delgado-Trejos, E., A. F. Quiceno-Manrique, J. I. Godino-Llorente, M. Blanco-Velasco, and G. Castellanos-Dominguez. Digital auscultation analysis for heart murmur detection. Ann. Biomed. Eng. 37(2):337–353, 2009.CrossRefPubMedGoogle Scholar
  17. 17.
    Djebbari, A. and F. Bereksi Reguig. Short-time Fourier transform analysis of the phonocardiogram signal. In: Proc. 7th IEEE International Conference on Electronics, Circuits and Systems, 2000, pp. 844–847.Google Scholar
  18. 18.
    Djurovic, I., and L. J. Stankovic. An algorithm for the Wigner distribution based instantaneous frequency estimation in a high noise environment. Signal Process. 84:631–643, 2004.CrossRefGoogle Scholar
  19. 19.
    El Segaier, M., O. Lilja, S. Lukkarinen, L. Sörnmo, R. Sepponen, and E. Pesonen. Computer-based detection and analysis of heart sound and murmur. Ann. Biomed. Eng. 33:937–942, 2005.CrossRefPubMedGoogle Scholar
  20. 20.
    Ergen, B., and Y. Tatar, Optimal continuous wavelet analysis of periodogram signals. In: Proc. of International Conference on Signal Processing, 2003, pp. 187–191.Google Scholar
  21. 21.
    Ergen, B., and Y. Tatar, Time–Frequency Analysis of Phonocardiogram. In: Proc. of 4th International Conference on Measurement, 2003, pp. 222.Google Scholar
  22. 22.
    Gajic, B., and K. K. Paliwal. Robust feature extraction using subband spectral centroid histograms. In: Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing, 2001, pp. 85–88.Google Scholar
  23. 23.
    Gill, D., N. Gavrieli, and N. Intrator. Detection and identification of heart sounds using homomorphic envelogram and self-organizing probabilistic model. In: Proc. Computers in Cardiology, 2005, pp. 957–960.Google Scholar
  24. 24.
    Groutage, D., and D. Bennink. A new matrix decomposition based on optimum transformation of the singular value decomposition basis sets yields principal features of time–frequency distributions. In: Proc. 10th IEEE Workshop on Statistical Signal and Array Processing, 2000, pp. 598–602.Google Scholar
  25. 25.
    Haghighi-Mood, A., and J. N. Torry. Time–frequency analysis of systolic murmurs. In: Proc. IEE Colloquium on Time–Frequency Analysis of Biomedical Signals, 1997, pp. 71–72.Google Scholar
  26. 26.
    Haibin, W., W. Jianqi, L. Guohua, Z. Guohui, and N. Ansheng. Application of adaptive time–frequency analysis in cardiac murmurs signal processing. In: Proc. 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2001, pp. 1896–1898.Google Scholar
  27. 27.
    Hamid, H., M. Mostefa, and B. Boualem. Time–frequency feature extraction of newborn EEG seizure using SVD-based techniques. EURASIP J. Appl. Signal Process. 2004(1):2544–2554, 2004.CrossRefGoogle Scholar
  28. 28.
    Hanley, J. A., and B. J. McNeil. A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology 148(3):839–843, 1983.PubMedGoogle Scholar
  29. 29.
    Hussain, Z. M., and B. Boashash. Design of time–frequency distributions for amplitude and IF estimation of multicomponent signals. In: Proc. of the 6th International Symposium on Signal Processing and Its Applications, 1st edn, 2001, pp. 339–342.Google Scholar
  30. 30.
    Jeharon, H., A. Seagar, and N. Seagar. Feature extraction from phonocardiogram for diagnosis based on expert system. In: Proc. 27th Annual International Conference of the Engineering in Medicine and Biology Society, edited by A. Seagar, 2005, pp. 5479–5482.Google Scholar
  31. 31.
    Kaiser, J. F. On a simple algorithm to calculate the ‘energy’ of a signal. In: Proc. International Conference on Acoustics, Speech, and Signal Processing, 1990, pp. 381–384.Google Scholar
  32. 32.
    Kamarulafizam, I., S. H. Salleh, J. M. Najeb, A. K. Ariff, and A. Chowdhury. Heart sound analysis using MFCC and time frequency distribution. In: 3rd Kuala Lumpur International Conference on Biomedical Engineering, 2006, pp. 402–405Google Scholar
  33. 33.
    Kikkawa, S., and M. Ishida. Number of degrees of freedom correlation times, and equivalent bandwidths of a random process. IEEE Trans. Inform. Theory 34:151–155, 1988.CrossRefGoogle Scholar
  34. 34.
    Kohler, B. U., C. Hennig, and R. Orglmeister. The principles of software QRS detection. IEEE Eng. Med. Biol. Mag. 21(1):42–57, 2002.CrossRefPubMedGoogle Scholar
  35. 35.
    Kouibia, A., and M. Pasadas. Approximation by shape preserving interpolation splines. Appl. Numerical Math. 37(3):271–288, 2001.CrossRefGoogle Scholar
  36. 36.
    Kumar, D., P. Carvalho, M. Antunes, J. Henriques, L. Eugenio, R. Schmidt, and J. Habetha. Detection of S1 and S2 heart sounds by high frequency signatures. In: Proc. 28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2006, pp. 1410–1416.Google Scholar
  37. 37.
    Kumar, D., R. Carvalho, M. Antunes, R. Gil, J. Henriques, and L. Eugenio. A new algorithm for detection of S1 and S2 heart sounds. In Proc. IEEE International Conference on Acoustics, Speech and Signal Processing, edited by R. Carvalho, 2006Google Scholar
  38. 38.
    Liang, H., and I. Nartimo. A feature extraction algorithm based on wavelet packet decomposition for heart sound signals. In: Proc. IEEE-SP International Symposium on Time–Frequency and Time-Scale Analysis, 1998, pp. 93–96.Google Scholar
  39. 39.
    Liang, H., S. Lukkarinen, and I. Hartimo. Heart sound segmentation algorithm based on heart sound envelogram. In: Proc. Computers in Cardiology, 1997, pp. 105–108.Google Scholar
  40. 40.
    Lima, C. S., and M. J. Cardoso. Phonocardiogram segmentation by using an hybrid RBF-HMM model. In: Proc. of the 5th IASTED International Conference on Biomedical Engineering, 2007, pp. 419–422.Google Scholar
  41. 41.
    Lima, C. S., and M. J. Cardoso. Phonocardiogram segmentation by using hidden Markov models. In: Proc. of the 5th IASTED International Conference on Biomedical Engineering, 2007, pp. 415–418.Google Scholar
  42. 42.
    Mahmoud, S. S., Z. M. Hussain, I. Cosic, and Q. Fang. Time–frequency analysis of normal and abnormal biological signals. Biomed. Signal Process. Control 1:33–43, 2006.CrossRefGoogle Scholar
  43. 43.
    Marchant, B. P. Time–frequency analysis for biosystems engineering. Biosyst. Eng. 85:261–281, 2003.CrossRefGoogle Scholar
  44. 44.
    Myint, W. W., and B. Dillard. An electronic stethoscope with diagnosis capability. In: Proc. 33rd Southeastern Symposium on System Theory, 2001, pp. 133–137.Google Scholar
  45. 45.
    Ning, T., and K. S. Hsieh. Delineation of systolic murmurs by autoregressive modelling. In: Proc. IEEE 21st Annual Northeast Bioengineering Conference, 1995, pp. 19–21.Google Scholar
  46. 46.
    Rankine, L., M. Mesbah, and B. Boashash. IF estimation for multicomponent signals using image processing techniques in the time–frequency domain. Signal Process. 87:1234–1250, 2007.CrossRefGoogle Scholar
  47. 47.
    Ricke, A. D., R. J. Povinelli, and M. T. Johnson. Automatic segmentation of heart sound signals using hidden Markov models. In: Proc. Computers in Cardiology, 2005, pp. 953–956.Google Scholar
  48. 48.
    Santos, M. A. R., and M. N. Souza. Detection of first and second cardiac sounds based on time frequency analysis. In: Proc. 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2001, pp. 1915–1918.Google Scholar
  49. 49.
    Sejdic, E., and J. J. Jiang. Comparative study of three time–frequency representations with applications to a novel correlation method. In: Proc. IEEE International Conference on Acoustics, Speech, and Signal Processing, 2004, pp. 633–636.Google Scholar
  50. 50.
    Sharif, Z., M. S. Zainal, A. Z. Sháameri, and S. H. S. Salleh. Analysis and classification of heart sounds and murmurs based on the instantaneous energy and frequency estimations. In: Proc. of TENCON, 2000, pp. 130–134.Google Scholar
  51. 51.
    Shino, H., H. Yoshida, H. Mizuta, and K. Yana. Phonocardiogram classification using time–frequency representation. In: Proc. 19th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, edited by H. Yoshida, 1997, pp. 1636–1637.Google Scholar
  52. 52.
    Sörnmo, L., and P. Laguna. Bioelectrical Signal Processing in Cardiac and Neurological Applications. Amsterdam: Elsevier Academic Press, 2005.Google Scholar
  53. 53.
    Sun, L., M. Shen, F. H. Y. Chan, and P. J. Beadle. Instantaneous frequency estimate of nonstationary phonocardiograph signals using Hilbert spectrum. In: Proc. 27th Annual International Conference of the Engineering in Medicine and Biology Society, 2005, pp. 7285–7288.Google Scholar
  54. 54.
    Sun, M., M. L. Scheuer, and R. J. Sclabassi. Decomposition of biomedical signals for enhancement of their time–frequency distributions. J. Franklin Inst. 337:453–467, 2000.CrossRefGoogle Scholar
  55. 55.
    Tavel, M. E., and H. Katz. Usefulness of a new sound spectral averaging technique to distinguish an innocent systolic murmur from that of aortic stenosis. Am. J. Cardiol. 95:902–904, 2005.CrossRefPubMedGoogle Scholar
  56. 56.
    Telatar, Z., and O. Erogul. Heart sounds modification for the diagnosis of cardiac disorders, In: Proc. of International Conference on Signal Processing, 1 edn., pp. 100–105, 2003Google Scholar
  57. 57.
    Tilkian, A., and M. Conover. Understanding heart sounds and murmurs: With an introduction to lung sounds. Philadelphia: W. B. Saunders Co., p. 358, 2001.Google Scholar
  58. 58.
    Tomasi, G., F. van den Berg, and C. Andersson. Correlation optimized warping and dynamic time warping as preprocessing methods for chromatographic data. J. Chemometr. 18:231–241, 2004.CrossRefGoogle Scholar
  59. 59.
    Tsipouras, M. G., and D. I. Fotiadis. Automatic arrhythmia detection based on time and time–frequency analysis of heart rate variability. Comput. Methods Progr. Biomed. 74:95–108, 2004.CrossRefGoogle Scholar
  60. 60.
    Turkoglu, I., A. Arslan, and E. Ilkay. An expert system for diagnosis of the heart valve diseases. Expert Syst. Appl. 23:229–236, 2002.CrossRefGoogle Scholar
  61. 61.
    Voss, A., J. Herold, R. Schroeder, F. Nasticzky, A. Mix, P. Ulrich, and T. Huebner. Diagnosis of aortic valve stenosis by correlation analysis of wavelet filtered heart sounds. In: Proc. 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2003, pp. 2873–2876.Google Scholar
  62. 62.
    Voss, A., A. Mix, and T. Hubner. Diagnosing aortic valve stenosis by parameter extraction of heart sound signals. Ann. Biomed. Eng. 33:1167–1174, 2005.CrossRefPubMedGoogle Scholar
  63. 63.
    Wang, P., Y. Kim, and C. B. Soh. Feature extraction based on Mel-scaled wavelet transform for heart sound analysis. In: Proc. 27th Annual International Conference of the Engineering in Medicine and Biology Society, 2005, pp. 7572–7575.Google Scholar
  64. 64.
    Wang, P., C. S. Lim, S. Chauhan, J. Y. Foo, and V. Anantharaman. Phonocardiographic signal analysis method using a modified hidden Markov model. Ann. Biomed. Eng. 35:367–374, 2007.CrossRefPubMedGoogle Scholar
  65. 65.
    White, P. R., W. B. Collis, and A. P. Salmon. Analyzing heart murmurs using time–frequency methods. In: Proc. IEEE-SP International Symposium on Time-Frequency and Time-Scale Analysis, 1996, pp. 385–388.Google Scholar
  66. 66.
    Yoshida, H., and S. Kikkawa. Information theoretic equivalent bandwidths of random processes and their applications. Methods Inf. Med. 46:110–116, 2007.PubMedGoogle Scholar
  67. 67.
    Yoshida, H., H. Shino, and K. Yana. Instantaneous frequency analysis of systolic murmur for phonocardiogram. In: Proc. 19th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 1997, pp. 1645–1647.Google Scholar

Copyright information

© Biomedical Engineering Society 2009

Authors and Affiliations

  • A. F. Quiceno-Manrique
    • 1
    Email author
  • J. I. Godino-Llorente
    • 2
  • M. Blanco-Velasco
    • 3
  • G. Castellanos-Dominguez
    • 1
  1. 1.Departamento de Ingeniería Eléctrica Electrónica y ComputaciónUniversidad Nacional de ColombiaCaldasColombia
  2. 2.Departamento de Ingeniería de Circuitos y SistemasUniversidad Politécnica de MadridMadridSpain
  3. 3.Departamento de Teoría de la Señal y ComunicacionesUniversidad de AlcaláAlcalá de HenaresSpain

Personalised recommendations