Abstract
This paper proposes a set of Deep Learning algorithms for classifying Phonocardiogram (PCG) scalogram. PCG signals contain valuable information about the heart health status, and they could help us in early detection and diagnosis of potential abnormalities. The system will classify into normal or abnormal categories, supported on reliable signal processing algorithms to automatically denoise and segment the sounds to improve the Deep Learning detection task. At the first stage, we denoised the PCG signal using a multi-resolution analysis based on the Discrete Wavelet Transform (DWT). At the second one, we segment automatically the sounds using an algorithm based on the Teager Energy Operator (TEO) and the autocorrelation. This is very important, because it is needed to select the S1 component related to the systole, and S2 component related to the diastole. Finally, scalogram images are obtained using Continuous Wavelet Transform (CWT). The classification task has been executed using the heart sounds from the 2016 PhysioNet/CinC Challenge database, and pretrained Convolutional Neural Networks (CNNs) ResNet152 and VGG16, achieving an accuracy of 91.19% and 90.75%, respectively. The results of our proposed model presents a good contribution to heart sounds classification area, in comparison with the state of the art accuracy which is 87%.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Alaskar, H., Alzhrani, N., Hussain, A., Almarshed, F.: The implementation of pretrained AlexNet on PCG classification. In: Huang, D.-S., Huang, Z.-K., Hussain, A. (eds.) ICIC 2019. LNCS (LNAI), vol. 11645, pp. 784–794. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-26766-7_71
Géron, A.: Hands-on machine learning with Scikit-Learn, Keras and TensorFlow: concepts, tools, and techniques to build intelligent systems (2019). https://www.oreilly.com/library/view/hands-on-machine-learning/9781492032632/
Benali Amjoud, A., Amrouch, M.: Convolutional neural networks backbones for object detection. In: El Moataz, A., Mammass, D., Mansouri, A., Nouboud, F. (eds.) ICISP 2020. LNCS, vol. 12119, pp. 282–289. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-51935-3_30
Bentley, P., Nordehn, G., Coimbra, M., Mannor, S.: The PASCAL Classifying Heart Sounds Challenge 2011 (CHSC2011) Results (2011). http://www.peterjbentley.com/heartchallenge/index.html
Chowdhury, T.H., Poudel, K.N., Hu, Y.: Time-frequency analysis, denoising, compression, segmentation, and classification of PCG signals. IEEE Access 8, 160882–160890 (2020). https://doi.org/10.1109/ACCESS.2020.3020806
He, Y., Li, W., Zhang, W., Zhang, S., Pi, X., Liu, H.: Research on segmentation and classification of heart sound signals based on deep learning. Appl. Sci. 11(2), 1–15 (2021). https://doi.org/10.3390/app11020651
Jain, P.K., Tiwari, A.K.: An adaptive thresholding method for the wavelet based denoising of phonocardiogram signal. Biomed. Signal Process. Control 38, 388–399 (2017). https://doi.org/10.1016/j.bspc.2017.07.002
Khan, A., Sohail, A., Zahoora, U., Qureshi, A.S.: A survey of the recent architectures of deep convolutional neural networks. Artif. Intell. Rev. 53(8), 5455–5516 (2020). https://doi.org/10.1007/s10462-020-09825-6
Khan, F.A., Abid, A., Khan, M.S.: Automatic heart sound classification from segmented/unsegmented phonocardiogram signals using time and frequency features. Physiol. Meas. 41(5), 55006 (2020). https://doi.org/10.1088/1361-6579/ab8770
Khan, N.M., Khan, M.S., Khan, G.M.: Automated Heart Sound Classification from Unsegmented Phonocardiogram Signals Using Time Frequency Features (2018). https://doi.org/10.5281/ZENODO.1340418. https://zenodo.org/record/1340418
Liu, C., et al.: An open access database for the evaluation of heart sound algorithms. Physiol. Meas. 37(12), 2181–2213 (2016). https://doi.org/10.1088/0967-3334/37/12/2181
Liu, Q., Wu, X., Ma, X.: An automatic segmentation method for heart sounds. BioMedical Eng. OnLine 17(1), 106 (2018). https://doi.org/10.1186/s12938-018-0538-9
Narváez, P., Gutierrez, S., Percybrooks, W.S.: Automatic segmentation and classification of heart sounds using modified empirical wavelet transform and power features. Appl. Sci. 10(14) (2020). https://doi.org/10.3390/app10144791
Nathalie Japkowicz, M.S.: Evaluating Learning Algorithms: A Classification Perspective (2011)
World Health Organization: A global brief on hypertension (2013). https://www.who.int/cardiovascular_diseases/publications/global_brief_hypertension/en/
World Health Organization: Cardiovascular diseases (CVDs) (2017). https://www.who.int/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds)
World Health Organization: WHO reveals leading causes of death and disability worldwide: 2000–2019 (2020). https://www.who.int/news/item/09-12-2020-who-reveals-leading-causes-of-death-and-disability-worldwide-2000-2019
Ramović, A., Bandić, L., Kevrić, J., Germović, E., Subasi, A.: Wavelet and teager energy operator (TEO) for heart sound processing and identification. In: Badnjevic, A. (ed.) CMBEBIH 2017, pp. 495–502. Springer, Singapore (2017). https://doi.org/10.1007/978-981-10-4166-2_76
Sabir, M.K.: PCG signal analysis using Teager energy operator & autocorrelation function. In: International Conference on Computer Medical Applications, ICCMA 2013 (2013). https://doi.org/10.1109/ICCMA.2013.6506176
Springer, D., Tarassenko, L., Clifford, G.: Logistic regression-HSMM-based heart sound segmentation. IEEE Trans. Biomed. Eng. 1 (2015). https://doi.org/10.1109/TBME.2015.2475278. http://ieeexplore.ieee.org/document/7234876/
WEISANG: Continuous Wavelet Transform (CWT) (2019). https://www.weisang.com/en/documentation/timefreqspectrumalgorithmscwt_en/
Acknowledgement
This work was partially funded by the University of Nariño and eVIDA group IT905-16 of the University of Deusto.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Gelpud, J., Castillo, S., Jojoa, M., Garcia-Zapirain, B., Achicanoy, W., Rodrigo, D. (2021). Deep Learning for Heart Sounds Classification Using Scalograms and Automatic Segmentation of PCG Signals. In: Rojas, I., Joya, G., Català, A. (eds) Advances in Computational Intelligence. IWANN 2021. Lecture Notes in Computer Science(), vol 12861. Springer, Cham. https://doi.org/10.1007/978-3-030-85030-2_48
Download citation
DOI: https://doi.org/10.1007/978-3-030-85030-2_48
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-85029-6
Online ISBN: 978-3-030-85030-2
eBook Packages: Computer ScienceComputer Science (R0)