Automated extraction of fine and coarse crackles by independent component analysis

  • Martín Emiliano Rodríguez-GarcíaEmail author
  • Sonia Charleston-Villalobos
  • Norma Castañeda-Villa
  • Aída Jiménez-González
  • Ramón González-Camarena
  • Ángel Tomás Aljama-Corrales
Original Paper


In this work further research was achieved on the separation by Independent Components (IC) of simulated abnormal breathing sounds (ABS) sources immersed in normal breathing sounds. This study considers only ABS discontinuous sounds, known as crackles, and includes fine and coarse types for both inspiratory and expiratory phases. Additionally, we develop a novel proposal to automated characterization of the IC associated with crackles. We analyzed the efficiency of three independent component analysis algorithms, i.e. FastICA, Infomax and TDSEP, through the Amari index, the signal to interference ratio, and the total relative distortion index. In the simulated multichannel signal scenarios, the performance indexes showed that Infomax is the best algorithm to solve the problem of blind source separation, supporting the results found in previous efforts. Finally, the presence of crackles in the IC obtained by Infomax was determined through their kurtosis and skewness, whereas the type of crackle was found by their characterization via the spectrogram of selected IC. Results indicate that the proposed methodology is able to adequately extract the crackle sources and identify the respiratory cycle phase in which they appear. Also, we managed to estimate the type and number of existing crackles in each source. In conclusion, our methodology can provide quantitative information on the clinical relevance of crackles in respiratory patients.


Blind source separation Crackles Kurtosis Skewness Spectrogram 



There is no funding source.

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Informed consent

Informed consent was obtained from all individual participants included in the study.


  1. 1.
    Castañeda-Villa N, Charleston-Villalobos S, González-Camarena R, Aljama-Corrales T. Assessment of ICA algorithms for the analysis of crackles sounds. Conf Proc IEEE Eng Med Biol Soc. 2012:605–8.Google Scholar
  2. 2.
    Reyes BA, Charleston-Villalobos S, González-Camarena R, Aljama-Corrales T. Assessment of time-frequency representation techniques for thoracic sounds analysis. Comput Methods Prog Biomed. 2014;114(3):276–90.CrossRefGoogle Scholar
  3. 3.
    Charleston-Villalobos S, Castañeda-Villa N, González-Camarena R, Mejía-Ávila M, Aljama-Corrales T. Automated clustering of independent components for discontinuous sounds thoracic imaging. ConfProc IEEE EngMedBiol Soc 2015;4126–9.Google Scholar
  4. 4.
    Charleston-Villalobos S, González-Camarena R, Chi-Lem G, Aljama-Corrales T. Crackle sounds analysis by empirical mode decomposition. IEEE Eng Med Biol Mag. 2007;26(1):40–7.CrossRefGoogle Scholar
  5. 5.
    Mantini D, Hild K, Alleva G, Comani S. Performance comparison of independent component analysis algorithms for fetal cardiac signal reconstruction: a study on synthetic fMCG data. Phys Med Biol. 2006;51(4):1033–46.CrossRefGoogle Scholar
  6. 6.
    Kamiya A, Murayama S, Kamiya H, Yamashiro T, Oshiro Y, Tanaka N. Kurtosis and skewness assessments of solid lung nodule density histograms: differentiating malignant from benign nodules on CT. Jpn J Radiol. 2014;32:14–21.CrossRefGoogle Scholar
  7. 7.
    Flandrin P. Time–frequency filtering based on spectrogram zeros. IEEE Signal Processing Letters. 2015;22(11):2137–41.CrossRefGoogle Scholar
  8. 8.
    Kiyokawa H, Greenberg M, Shirota K, Pasterkamp H. Auditory detection of simulated crackles in breath sounds. Chest. 2001;119(6):1886–92.Google Scholar
  9. 9.
    Quandt V, Pacola E, Pichorim S, Gamba H, Sovierzoski M. Pulmonary crackle characterization: approaches in the use of discrete wavelet transform regarding border effect, mother-wavelet selection, and subband reduction. Res Biomed Eng. 2015.Google Scholar
  10. 10.
    Nordhausen K, Ollila E, Oja H. On the performance indices of ICA and blind source separation. IEEE 12th international workshop on signal processing advances in wireless Communications. 2011;486–490.Google Scholar
  11. 11.
    Gribonval R, Benaroya L, Vincent E, Févotte C. Proposals for performance measurement in source separation. ICA. 2003;2003:763–8.Google Scholar

Copyright information

© IUPESM and Springer-Verlag GmbH Germany, part of Springer Nature 2019

Authors and Affiliations

  • Martín Emiliano Rodríguez-García
    • 1
    Email author
  • Sonia Charleston-Villalobos
    • 1
  • Norma Castañeda-Villa
    • 1
  • Aída Jiménez-González
    • 1
  • Ramón González-Camarena
    • 2
  • Ángel Tomás Aljama-Corrales
    • 1
  1. 1.Electrical Engineering DepartmentMetropolitan Autonomous UniversityMexico CityMexico
  2. 2.Health Sciences DepartmentMetropolitan Autonomous UniversityMexico CityMexico

Personalised recommendations