Abstract
In this research, we propose a method for recovering heart sound signals that comprises the long short-term memory prediction model based on the recurrent neuron network architecture. The complete heart sound signal is used to implement a prediction model to recover damaged or incomplete heart sound signals. Root mean square errors (RMSEs) and Pearson’s correlation coefficients are used for numerical evaluation. The signals of 13 out of 15 subjects are considerably improved, with the RMSE being as low as 0.03 ± 0.04. Using the Pearson correlation coefficients to estimate the degree of signal recovery, the highest coefficient of correlation between the original and recovered time-domain waveforms is 0.93, and that between the corresponding spectra is 0.967. Waveforms and spectra are used to compare the results graphically. The recovered signal more closely fits the original signal than the interfered signal. Additionally, excess frequency components in the recovered spectra are found to be filtered out and important features retained. Thus, the proposed method not only recovers incomplete or disturbed signals but also has the effect of a filter.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11227-019-03096-x/MediaObjects/11227_2019_3096_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11227-019-03096-x/MediaObjects/11227_2019_3096_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11227-019-03096-x/MediaObjects/11227_2019_3096_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11227-019-03096-x/MediaObjects/11227_2019_3096_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11227-019-03096-x/MediaObjects/11227_2019_3096_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11227-019-03096-x/MediaObjects/11227_2019_3096_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11227-019-03096-x/MediaObjects/11227_2019_3096_Fig7_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11227-019-03096-x/MediaObjects/11227_2019_3096_Fig8_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11227-019-03096-x/MediaObjects/11227_2019_3096_Fig9_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11227-019-03096-x/MediaObjects/11227_2019_3096_Fig10_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11227-019-03096-x/MediaObjects/11227_2019_3096_Fig11_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs11227-019-03096-x/MediaObjects/11227_2019_3096_Fig12_HTML.png)
Similar content being viewed by others
References
Bertrand CA, Milne IG, Hornick R (1956) A study of heart sounds and murmurs by direct heart recordings. Circulation 13(1):49–57
Kono T, Rosman H, Alam M, Stein PD, Sabbah HN (1993) Hemodynamic correlates of the third heart sound during the evolution of chronic heart failure. J Am Coll Cardiol 21(2):419–423
Hada Y, Amano K et al (1986) Noninvasive study of the presystolic component of the first heart sound in mitral stenosis. J Am Coll Cardiol 7(1):43–50
Farrar MW, Engel PJ, Eppert D, Plummer S (1985) Late systolic click from isolated tricuspid valve prolapse simulating paradoxical splitting of the second heart sound. J Am Coll Cardiol 5(3):793–796
Ishimitsu T, Smith D, Berko B, Craige E (1985) Origin of the third heart sound: comparison of ventricular wall dynamics in hyperdynamic and hypodynamic types. J Am Coll Cardiol 5:268–272
Ahlstrom C, Hult P, Rask P, Karlsson J-E, Nylander E, Dahlström U, Ask P (2006) Feature extraction for systolic heart murmur classification. Ann Biomed Eng 34(11):1666–1677
Amiri AM, Armano G (2013) Heart sound analysis for diagnosis of heart diseases in newborns. APCBEE Procedia 7:109–116
Danford DA (2004) Heart murmur in child. Turner White Communications Inc, Wayne
Messner E, Zöhrer M, Pernkopf F (2018) Heart sound segmentation—an event detection approach using deep recurrent neural networks. Trans Biomed Eng 65(9):1964–1974
Dwivedi AK, Imtiaz SA, Rodriguez-Villegas E (2019) Algorithms for automatic analysis and classification of heart sounds—a systematic review. Access 7:8316–8345
Choudhary T, Sharma LN, Bhuyan MK (2018) Heart sound extraction from sternal seismocardiographic signal. Signal Process Lett 25(4):482–486
Mishra M, Banerjee S, Thomas DC, Dutta S, Mukherjee A (2018) Detection of third heart sound using variational mode decomposition. Trans Instrum Meas 67(7):1713–1721
Oliveira J, Renna F, Mantadelis T, Coimbra M (2019) Adaptive sojourn time HSMM for heart sound segmentation. J Biomed Health Inform 23(2):642–649
Babu KA, Ramkumar B, Manikandan MS (2018) Automatic identification of S1 and S2 heart sounds using simultaneous PCG and PPG recordings. Sens J 18(22):9430–9440
Latif S, Usman M, Rana R, Qadir J (2018) Phonocardiographic sensing using deep learning for abnormal heartbeat detection. Sens J 18(22):9393–9400
Mondal A, Saxena I, Tang H, Banerjee P (2018) A noise reduction technique based on nonlinear kernel function for heart sound analysis. J Biomed Health Inform 22(3):775–784
Elamaran V, Arunkumar N, Hussein AF, Solarte M, Ramirez-Gonzalez G (2018) Spectral fault recovery analysis revisited with normal and abnormal heart sound signals. Access 6:62874–62879
Emmanouilidou D, McCollum ED, Park DE, Elhilali M (2018) Computerized lung sound screening for pediatric auscultation in noisy field environments. Trans Biomed Eng 65(7):1564–1574
Sharma P, Imtiaz SA, Rodriguez-Villegas E (2019) An algorithm for heart rate extraction from acoustic recordings at the neck. Trans Biomed Eng 66(1):246–256
Nivitha Varghees V, Ramachandran KI, Soman KP (2018) Wavelet-based fundamental heart sound recognition method using morphological and interval features. Healthc Technol Lett 5(3):81–87
Dominguez-Morales JP, Jimenez-Fernandez AF, Dominguez-Morales MJ, Jimenez-Moreno G (2018) Deep neural networks for the recognition and classification of heart murmurs using neuromorphic auditory sensors. Trans Biomed Circuits Syst 12(1):24–34
Siegelmann HT, Sontag ED (1992) On the computational power of neural nets. In: COLT’92. ACM, pp 440–449
Graves A, Liwicki M, Fernández S, Bertolami R, Bunke H, Schmidhuber J (2009) A novel connectionist system for improved unconstrained handwriting recognition. Trans Pattern Anal Mach Intell 31(5):855–868
Sak H, Senior A, Beaufays F (2014) Long short-term memory recurrent neural network architectures for large scale acoustic modeling. ar**v:1402.1128
Li X, Wu X (2014) Constructing long short-term memory based deep recurrent neural networks for large vocabulary speech recognition. ar**v:1410.4281
Jahani B, Mohammadi B (2019) A comparison between the application of empirical and ANN methods for estimation of daily global solar radiation in Iran. Theor Appl Climatol 137(1–2):1257–1269
Moazenzadeha R, Mohammadib B (2019) Assessment of bio-inspired metaheuristic optimisation algorithms for estimating soil temperature. Geoderma 353(1):152–171
Mohammadi B (2019) Letter to the editor “Estimation of sodium adsorption ratio indicator using data mining methods: a case study in Urmia Lake basin, Iran” by Mohammad Taghi Sattari, Arya Farkhondeh, and John Patrick Abraham. Environ Sci Pollut Res 26(10):10439–10440
Mohammadi B (2019) Letter to the editor “Design of an integrated climatic assessment indicator (ICAI) for wheat production: a case study in Jiangsu Province, China” by **angying Xu, ** Gao, **nkai Zhu, Wenshan Guo, **feng Ding, Chunyan Li, Min Zhu, Xuanwei Wu. Ecol Ind 103:493
Acknowledgements
This work was supported in part by the “Allied Advanced Intelligent Biomedical Research Center, STUST” from Higher Education Sprout Project, Ministry of Education, Taiwan and in part by the Small Business Innovation Research (SBIR) of Taiwan under No.1Z1070773.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Wang, ZH., Horng, GJ., Hsu, TH. et al. Heart sound signal recovery based on time series signal prediction using a recurrent neural network in the long short-term memory model. J Supercomput 76, 8373–8390 (2020). https://doi.org/10.1007/s11227-019-03096-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-019-03096-x