Abstract
This paper proposes a robust and fully automated respiratory phase segmentation method using single channel tracheal breath sounds (TBS) recordings of different types. The estimated number of respiratory segments in a TBS signal is firstly obtained based on noise estimation and nonlinear mapping. Respiratory phase boundaries are then located through the generations of multi-population genetic algorithm by introducing a new evaluation function based on sample entropy (SampEn) and a heterogeneity measure. The performance of the proposed method is analyzed for single channel TBS recordings of various types. An overall respiratory phase segmentation accuracy is found to be 12 ± 5 ms for normal TBS and 21 ± 9 ms for adventitious sounds. The results show the robustness and effectiveness of the proposed segmentation method. The proposed method has been a successful attempt to solve the clinical application challenge faced by the existing phase segmentation methods in terms of respiratory dysfunctions.
Similar content being viewed by others
References
Abeyratne UR, Karunajeewa AS, Hukins C (2007) Mixed-phase modeling in snore sound analysis. Med Biol Eng Comput 45(8):791–806
Ashkanazi J, Silverberg P, Foster R, Hyman A, Milic-Emili J, Kinney J (1980) Effects of respiratory apparatus on breathing pattern. J Appl Physiol 48:577–580
Berouti M, Schwartz R, Makhoul J (1979) Enhancement of speech corrupted by acoustic noise, vol 4. Proceedings of 4th IEEE ICASSP conference, pp 208–211
Cam SL, Collet Ch, Salzenstein F (2008) Acoustical respiratory signal analysis and phase detection. Proceedings of 33rd IEEE ICASSP conference, pp 3629–3632
Chen XN, Solomon IC, Chon KH (2005) Comparison of the use of approximate entropy and sample entropy: applications to neural respiratory signal. Proceedings of 27th IEEE EMBS Conference, pp 4212–4215
Chipperfield A, Fleming P, Pohlheim H, Fonseca C (1995) Genetic algorithm toolbox. Department of Automatic Control and Systems Engineering, University of Sheffield, UK
Coley DA (2001) An Introduction to genetic algorithms for scientists and engineers. World Scientific, New Jersy
Corté S, Jané R, Fiz JA, Morera J (2005) Monitoring of wheeze duration during spontaneous respiration in asthmatic patients. Proceedings of 27th IEEE EMBS Conference
Erkelens JS, Heusdens R (2008) Tracking of nonstationary noise based on data-driven recursive noise power estimation. IEEE Trans Audio Speech Lang Process 16(6):1112–1123
Field DJ (1987) Relations between the statistics of natural images and the response properties of cortical cells. J Opt Soc Am 4:2379–2394
Goldberg DE (1989) Genetic algorithm in search, optimization and machine learning. Addision-Wesley, USA
Huang Y, Benesty J, Chen JD (2006) Acoustic MIMO signal processing. Springer, Berlin
Huang Q, Dom B (1995) Quantitative methods of evaluating image segmentation. IEEE Proc Int Conf Image Process 3:53–56
Hult P, Fjällbrant T, Wranne B, Engdahl O, Ask P (2004) An improved bioacoustic method for monitoring of respiration. Tech Health Care, pp 323–332
Kulkas A, Huupponen E, Virkkala J, Tenhunen M, Saastamoinen A, Rauhala E, Himanen SL (2009) New tracheal sound feature for apnoea analysis. Med Biol Eng Comput 47(4):405–412
Larsen ER (2004) Audio bandwidth extension: application of psychoacoustics, signal processing and loudspeaker design. Wiley, New York
Lehrer S (2002) Understanding lung sounds, audio CD. Saunders, Philadelphia
Marshall S, Sicuranza GL (2006) Advances in nonlinear signal and image processing. Hindawi Publishing Corporation
Meslier N, Charbonneau G, Racineux JL (1995) Wheezes. Eur Respir J 8:1942–1948
Mukhopadhyay S, Ray GC (1998) A new interpretation of nonlinear energy operator and its efficacy in spike detection. IEEE Trans Biomed Eng 45(2):180–187
Nongpiur RC (2008) Impulse noise removal in speech using wavelets, pp 1593–1596
Oppenheim AV, Schafer RW (1999) Discrete-time signal processing. Prentice-Hall, USA
Pincus SM (1995) Approximate entropy (ApEn) as a complexity measure. Chaos 5:110–117
Qi JG, Burns GR, Harrison DK (2000) The application of parallel multipopulation genetic algorithms to dynamic job-shop scheduling. Int J Adv Manufacturing Tech 16(8):609–615
Richman JS, Moorman JR (2000) Physiological time-series analysis using approximate entropy and sample entropy. Am J Physiol Heart Circ Physiol 278(6):H2039–2049
Schatzman M (2002) Numerical analysis: a mathematical introduction. Clarendon Press, Oxford
Shao Y, Chang CH (2007) A generalized time-frequency subtraction method for robust speech enhancement based on wavelet filter banks modeling of human auditory system. IEEE Trans Sys Man Cyber Part B 37(4):877–889
Sierra G, Telfort V, Popov B, Durand LG, Agarwal R, Lanzo V (2004) Monitoring respiratory rate based on tracheal sounds. first experiences. Proceedings of 26th IEEE EMBS Conference, pp 317– 320
Sovijärvi ARA, Vanderschoot J, Eavis JR (2000) Standardization of computerized respiratory sound analysis. Eur Respir Rev 10(77):585–649
Tang KS, Man KF, Kwong S, He Q (1996) Genetic algorithms and their applications. Signal Process Maga IEEE 13(6):22–37
Taplidou SA, Hadjileontiadis LJ (2007) Nonlinear analysis of wheezes using wavelet bicoherence. Comput Biol Med 37:563–570
Wilkins RL, Hodgkin JE, Lopez B (2004) Fundamentals of lung and heart sounds, audio CD. Mosby, USA
Yadollahi A, Moussavi Z (2006) A robust method for estimating respiratory flow using tracheal sounds entropy. IEEE Trans Biomed Eng 53(4):662–668
Yap YL, Moussavi Z (2008) Respiratory onset detection using variance fractal dimension. Biomed Signal Process Control, pp 181–191
Yildirim I, Ansari R, Moussavi Z (2008) Automated resiratory phase and onset detection using only chest sound signal. Proceedings of 30th IEEE EMBS Conference, pp 2578–2581
Acknowledgments
The authors gratefully acknowledge the contribution of National University Hospital, especially Dr. Irene Melinda Louis, for their support in data collection and identification. The authors are also grateful to the Reviewers and the Editor for their valuable comments and suggestions that help to improve the paper significantly.
Author information
Authors and Affiliations
Corresponding author
Appendix 1
Appendix 1
SampEn(m, r, N) can be calculated as below:
For an input signal U of length N, {U(p):1 ≤ p ≤ N} forms the N−m + 1 vectors x m (q) for {q|1 ≤ q ≤ N−m + 1}, where x m (q) = {U(q + b):0 ≤ b ≤ m−1} is the vector of m data points from U(q) to U(q + m−1). In this context, only the first N−m vectors of length m are considered to ensure that, x m (q) and x m+1(q) are defined for 1 ≤ q ≤ N−m. Let B m(r) be the probability that two sequences will match for m points and A m(r) is the probability that two sequences will match for m + 1 points. B m q (r) is defined as (N−m−1)−1 times the numbers of vectors x m (p) within r of x m (q), where 1 ≤ p ≤ N−m, and p ≠ q to exclude self-matches. Then B m(r) is defined as
Similarly, A m q (r) is defined as (N−m−1)−1 times the numbers of vectors x m+1(p) within r of x m+1(q), where 1 ≤ p ≤ N−m and p ≠ q. Then set A m(r) as
Finally, sample entropy (SampEn) is calculated by
Rights and permissions
About this article
Cite this article
Jin, F., Sattar, F. & Goh, D.Y.T. An acoustical respiratory phase segmentation algorithm using genetic approach. Med Biol Eng Comput 47, 941–953 (2009). https://doi.org/10.1007/s11517-009-0518-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11517-009-0518-0