Skip to main content

Feature Extraction From Parametric Time–Frequency Representations for Heart Murmur Detection

Abstract

The detection of murmurs from phonocardiographic recordings is an interesting problem that has been addressed before using a wide variety of techniques. In this context, this article explores the capabilities of an enhanced time–frequency representation (TFR) based on a time-varying autoregressive model. The parametric technique is used to compute the TFR of the signal, which serves as a complete characterization of the process. Parametric TFRs contain a large quantity of data, including redundant and irrelevant information. In order to extract the most relevant features from TFRs, two specific approaches for dimensionality reduction are presented: feature extraction by linear decomposition, and tiling partition of the tf plane. In the first approach, the feature extraction was carried out by means of eigenplane-based PCA and PLS techniques. Likewise, a regular partition and a refined Quadtree partition of the tf plane were tested for the tiled-TFR approach. As a result, the feature extraction methodology presented, which searches for the most relevant information immersed on the TFR, has demonstrated to be very effective. The features extracted were used to feed a simple k-nn classifier. The experiments were carried out using 45 phonocardiographic recordings (26 normal and 19 records with murmurs), segmented to extract 548 representative individual beats. The results using these methods point out that better accuracy and flexibility can be accomplished to represent non-stationary PCG signals, showing evidences of improvement with respect to other approaches found in the literature. The best accuracy obtained was 99.06 ± 0.06%, evidencing high performance and stability. Because of its effectiveness and simplicity of implementation, the proposed methodology can be used as a simple diagnostic tool for primary health-care purposes.

This is a preview of subscription content, access via your institution.

FIGURE 1
FIGURE 2
FIGURE 3
FIGURE 4
FIGURE 5
FIGURE 6
FIGURE 7
FIGURE 8
FIGURE 9
FIGURE 10
FIGURE 11
FIGURE 12
FIGURE 13
FIGURE 14
FIGURE 15

Notes

  1. S1 implies the closing of the tricuspid and mitral valves immediately preceding the systole, whereas S2 corresponds to the closing of the aortic and pulmonary valves at the end of systole.

Abbreviations

2D-PCA:

Two-dimensional PCA

AR:

Autoregressive

BIC:

Bayesian information criterion

CWT:

Continuous wavelet transform

ECG:

Electrocardiogram

HS:

Heart sound

k-nn:

k-nearest neighbors

LS-TVAR:

Least-squares TVAR

PCA:

Principal component analysis

PCG:

Phonocardiogram

PLS:

Partial least squares

SNR:

Signal-to-noise ratio

tf:

Time–frequency

TFR:

Time–frequency representation

TVAR:

Time-varying autoregressive

WVD:

Wigner–Ville distribution

References

  1. Abramovich, Y., N. Spencer, and M. Turley. Order estimation and discrimination between stationary and time-varying (TVAR) autoregressive models. IEEE Trans. Signal Process. 55(6):2861–2876, 2007.

    Article  Google Scholar 

  2. Ahlstrom, C., P. Hult, P. Rask, J. Karlsson, E. Nylander, U. Dahlstrom, and P. Ask. Feature extraction for systolic heart murmur classification. Ann. Biomed. Eng. 34:1666–1677, 2006.

    Article  PubMed  Google Scholar 

  3. Avendano-Valencia, D., F. Martinez-Tabares, D. Acosta-Medina, I. Godino-Llorente, and G. Castellanos-Dominguez. TFR-based feature extraction using PCA Approaches for discrimination of heart murmurs. Proceedings of the 31th IEEE EMBS Annual International Conference (EMBC’09), 2009.

  4. Barker, M., and W. Rayens. Partial least squares for discrimination. J. Chemomet. 17(3):166–173, 2003.

    Article  CAS  Google Scholar 

  5. Bernat, E., W. Williams, and W. Gehring. Decomposing ERP time–frequency energy using PCA. Clin. Neurophysiol. 116:1314–1334, 2005.

    Article  PubMed  Google Scholar 

  6. Cassidy, M., and W. Penny. Bayesian nonstationary autoregressive models for biomedical signal analysis. IEEE Trans. Biomed. Eng. 49(10):1142–1152, 2002.

    Article  PubMed  Google Scholar 

  7. Cerutti, S., A. Bianchi, and L. Mainardi. Advanced spectral methods for detecting dynamic behaviour. Auton. Neurosci.: Basic Clin. 90(1):3–12, 2001.

    Article  CAS  Google Scholar 

  8. Delgado-Trejos, E., A. Quiceno-Manrique, J. Godino-Llorente, M. Blanco-Velasco, and G. Castellanos-Dominguez. Digital auscultation analysis for heart murmur detection. Ann. Biomed. Eng. 37(2):337–353, 2009.

    Article  PubMed  Google Scholar 

  9. Deng, J., J. Yao, J. Dewald, and P. Julius. Classification of the intention to generate a shoulder versus elbow torque by means of a time frequency synthesized spatial patterns BCI algorithm. J. Neural Eng. 2(4):131–138, 2005.

    Google Scholar 

  10. El-Segaier, M., O. Lilja, O. Lukkarinen, L. Sörnmo, R. Sepponen, and E. Pesonen. Computer-based detection and analysis of heart sound and murmur. Ann. Biomed. Eng. 33(7):937–942, 2005.

    Article  CAS  PubMed  Google Scholar 

  11. Englehart, K., B. Hudgins, P. Parker, and M. Stevenson. Classification of the myoelectric signal using time-frequency based representations. Med. Eng. Phys. 21(6):431–438, 1999.

    Article  CAS  PubMed  Google Scholar 

  12. Güler, I., M. Kiymik, and F. Güler. Order determination in autoregressive modeling of diastolic heart sounds. J. Med. Syst. 20(1):11–17, 1995.

    Article  Google Scholar 

  13. Kaipio, J., and M. Juntunen. Deterministic regression smoothness priors TVAR modelling. Proc. IEEE ICASSP 99, 1999, 1693–1696.

  14. Kanai, H., N. Chubachi, and Y. Koiwa. A time-varying AR modeling of heart wall vibration. In: Proceedings on International Conference of the Acoustics, Speech, and Signal Processing, ICASSP 95, edited by EEE Computer Society, 1995, pp. 941–944.

  15. Marchant, B. Time-frequency analysis for biosystems engineering. Biosyst. Eng. 85(3):261–281, 2003.

    Article  Google Scholar 

  16. Nandagopal, D., J. Mazumbar, and R. Bogner. Spectral analysis of second heart sound in children by selective linear prediction coding. Med. Biol. Eng. Comput. 22:229–239, 1985.

    Google Scholar 

  17. Poulimenos, A., and S. Fassois. Parametric time-domain methods for non-stationary random vibration modelling and analysis—a critical survey and comparison. Mech. Syst. Signal Process. 20(4):763–816, 2006.

    Article  Google Scholar 

  18. Quiceno-Manrique, A. F., J. I. Godino-Llorente, M. Blanco-Velasco, and G. Castellanos-Domínguez. Selection of dynamic features based on time-frequency representations for heart murmur detection from phonocardiographic signals. Ann. Biomed. Eng. 38(1):118–137, 2009.

    Article  PubMed  Google Scholar 

  19. Sejdic, E., I. Djurovic, and J. Jiang. Time–frequency feature representation using energy concentration: an overview of recent advances. Digital Signal Process. 19(1):153–183, 2009.

    Article  Google Scholar 

  20. Sullivan, G., and R. Baker. Efficient quadtree coding of images and video. IEEE Trans. Image. Process. 3(3):327–331, 1994.

    Article  CAS  PubMed  Google Scholar 

  21. Tarvainen, M., J. Hiltunen, P. Ranta-aho, and P. Karjalainen. Estimation of nonstationary EEG with Kalman smoother approach: an application to event-related synchronization. IEEE Trans. Biomed. Eng. 51(3):516–524, 2004.

    Article  PubMed  Google Scholar 

  22. Tzallas, A., M. Tsipouras, and D. Fotiadis. Automatic seizure detection based on time-frequency analysis and artificial neural networks. Comput. Intell. Neurosci. 2007:1–13, 2007.

    Article  Google Scholar 

  23. Wang, P., C. S. Lim, S. Chauhan, J. Yong, A. Foo, and V. Anantharaman. Phonocardiographic signal analysis method using a modified hidden Markov model. Ann. Biomed. Eng. 35(3):367–374, 2006.

    Article  PubMed  Google Scholar 

  24. Yang, J., D. Zhang, A. Frangi, and J. Yang. Two-dimensional PCA: a new approach to appearance-based face representation and recognition. IEEE Trans. Pattern Anal. Mach. Intell. 26(1):131–137, 2004.

    Article  PubMed  Google Scholar 

Download references

Acknowledgments

The authors would like to acknowledge Dr. Ana Maria Matijasevic and Dr. Guillermo Agudelo who are working with Universidad de Caldas for organizing the acquisition of the PCG data. This research was carried out under grants: “Centro de Investigación e Innovación de Excelencia ARTICA,” funded by COLCIENCIAS; and TEC2006-12887-C02 from the Ministry of Science and Technology of Spain.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to L. D. Avendaño-Valencia.

Additional information

Associate Editor Ioannis A. Kakadiaris oversaw the review of this article.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Avendaño-Valencia, L.D., Godino-Llorente, J.I., Blanco-Velasco, M. et al. Feature Extraction From Parametric Time–Frequency Representations for Heart Murmur Detection. Ann Biomed Eng 38, 2716–2732 (2010). https://doi.org/10.1007/s10439-010-0077-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10439-010-0077-4

Keywords

  • Heart sounds
  • Feature extraction
  • Time–frequency representation
  • Time-varying autoregressive model
  • Murmur detection