MW-Net: A Deep Framework for Segmentation and Detection of S1 and S2 Heart Sounds

  • Authors

    • Madhwendra Nath Department of Electronics and Communication Engineering, National Institute of Technology Patna, Patna-800005, India
    • Subodh Srivastava Department of Electronics and Communication Engineering, National Institute of Technology Patna, Patna-800005, India https://orcid.org/0000-0001-6194-429X
    https://doi.org/10.14419/k0nqfx70

    Received date: October 7, 2025

    Accepted date: October 12, 2025

    Published date: November 2, 2025

  • Complex Diffusion, Mish, Synchro-Squeezed Wavelet Transforms, Random Search, Unsharp Masking, W-net

  • Abstract

    This paper proposes a Mish W-Net deep framework for precise segmentation of the primary heart sounds, S1 and S2, from PCG signals. The proposed deep framework incorporates a synchro-squeezed wavelet transform for signal-to-image conversion. Additionally, a random search-optimized complex diffusion unsharp masking is proposed to mitigate the issues of noise and image quality after the signal-to-noise conversion. Besides this, to achieve self-regularizing feedback learning, the Mish activation function is employed at the architectural level of the proposed Mish W-Net. The effectiveness of the proposed deep framework is evaluated on the publicly available PhysioNet/Computing in Cardiology PCG dataset. Comparative qualitative and quantitative assessment outperforms the existing methodologies. The proposed method achieves accuracy, precision, recall, F1-score, and intersection over union of 98.28%, 98.77%, 99.50%, 99.13%, and 98.28%. It shows an effective framework that offers precise segmentation and detection of heart sounds.

  • References

    1. T. E. Chen et al., “S1 and S2 heart sound recognition using deep neural networks,” IEEE Trans. Biomed. Eng., vol. 64, no. 2, pp. 372–380, 2017, doi: 10.1109/TBME.2016.2559800.
    2. P. Joseph et al., “cardiovascular disease in the Americas: the epidemiology of cardiovascular disease and its risk factors,” Lancet Reg. Heal. - Am., vol. 42, p. 100960, 2025, doi: 10.1016/j.lana.2024.100960.
    3. M. D. Teja and G. M. Rayalu, “Hybrid time series and machine learning models for forecasting cardiovascular mortality in India: an age specific analysis,” BMC Public Health, vol. 25, no. 1, 2025, doi: 10.1186/s12889-025-23318-7.
    4. T. H. Chowdhury, K. N. Poudel, and Y. Hu, “Time-Frequency Analysis, Denoising, Compression, Segmentation, and Classification of PCG Sig-nals,” IEEE Access, vol. 8, pp. 160882–160890, 2020, doi: 10.1109/ACCESS.2020.3020806.
    5. S. A. Alali et al., “Optimized CNN-based denoising strategy for enhancing longitudinal monitoring of heart failure,” Comput. Biol. Med., vol. 184, no. November 2024, p. 109430, 2025, doi: 10.1016/j.compbiomed.2024.109430.
    6. M. Nath, S. Srivastava, N. Kulshrestha, and D. Singh, “Detection and localization of S1 and S2 heart sounds by 3rd order normalized average Shannon energy envelope algorithm,” Proc. Inst. Mech. Eng. Part H J. Eng. Med., vol. 235, no. 6, pp. 615–624, 2021.
    7. Y. Jang et al., “Fully Convolutional Hybrid Fusion Network with Heterogeneous Representations for Identification of S1 and S2 from Phonocardi-ogram,” IEEE J. Biomed. Heal. Informatics, vol. 28, no. 12, pp. 7151–7163, 2024, doi: 10.1109/JBHI.2024.3431028.
    8. M. Nath and S. Srivastava, “4th Order Shannon Energy Envelope Approach for Localization of S1 and S2 for Early-Stage Detection of Heart Valve Dysfunction,” Trait. du Signal, vol. 40, no. 2, pp. 479–490, 2023, doi: 10.18280/ts.400207.
    9. Y. Chen, Y. Sun, J. Lv, B. Jia, and X. Huang, “End-to-end heart sound segmentation using deep convolutional recurrent network,” Complex Intell. Syst., vol. 7, no. 4, pp. 2103–2117, 2021, doi: 10.1007/s40747-021-00325-w.
    10. C. Park et al., “Enhancement of phonocardiogram segmentation using convolutional neural networks with Fourier transform module,” Biomed. Eng. Lett., vol. 15, no. 2, pp. 401–413, 2025, doi: 10.1007/s13534-025-00458-8.
    11. Q. Liu, X. Wu, and X. Ma, “An automatic segmentation method for heart sounds,” Biomed. Eng. Online, vol. 17, no. 1, pp. 1–22, 2018, doi: 10.1186/s12938-018-0538-9.
    12. P. Xiao and K. Wang, “Segmentation of Heart Sound Signals Using Improved Hilbert Transform and Wavelet Packet Transform,” Circuits, Syst. Signal Process., vol. 44, no. 7, pp. 4752–4773, 2025, doi: 10.1007/s00034-025-03000-4.
    13. F. Renna, J. Oliveira, and M. T. Coimbra, “Deep Convolutional Neural Networks for Heart Sound Segmentation,” IEEE J. Biomed. Heal. Infor-matics, vol. 23, no. 6, pp. 2435–2445, 2019, doi: 10.1109/JBHI.2019.2894222.
    14. E. Messner, M. Zöhrer, and F. Pernkopf, “Heart sound segmentation - An event detection approach using deep recurrent neural networks,” IEEE Trans. Biomed. Eng., vol. 65, no. 9, pp. 1964–1974, 2018, doi: 10.1109/TBME.2018.2843258.
    15. T. Fernando, H. Ghaemmaghami, S. Denman, S. Sridharan, N. Hussain, and C. Fookes, “Heart Sound Segmentation Using Bidirectional LSTMs with Attention,” IEEE J. Biomed. Heal. Informatics, vol. 24, no. 6, pp. 1601–1609, 2020, doi: 10.1109/JBHI.2019.2949516.
    16. A. Mukherjee, R. Banerjee, and A. Ghose, “A Novel U-Net Architecture for Denoising of Real-world Noise Corrupted Phonocardiogram Signal,” 2023, [Online]. Available: http://arxiv.org/abs/2310.00216.
    17. V. M. Venegas et al., “Automated Phonocardiogram Segmentation a 1D U-Net Convolutional Neural Network : A Binary Approach,” pp. 0–12, 2025, doi: 10.20944/preprints202501.2087.v1.
    18. G. D. Clifford et al., “Classification of normal/abnormal heart sound recordings: The PhysioNet/Computing in Cardiology Challenge 2016,” Com-put. Cardiol. (2010)., vol. 43, pp. 609–612, 2016, doi: 10.22489/cinc.2016.179-154.
    19. I. Daubechies, J. Lu, and H.-T. Wu, “Synchrosqueezed Wavelet Transforms: a Tool for Empirical Mode Decomposition,” pp. 1–23, 2009, [Online]. Available: http://arxiv.org/abs/0912.2437.
    20. I. W. Selesnick and C. Sidney Burrus, “Generalized digital butterworth filter design,” IEEE Trans. Signal Process., vol. 46, no. 6, pp. 1688–1694, 1998, doi: 10.1109/78.678493.
    21. R. Srivastava and S. Srivastava, “Restoration of Poisson noise corrupted digital images with nonlinear PDE based filters along with the choice of regularization parameter estimation,” Pattern Recognit. Lett., vol. 34, no. 10, pp. 1175–1185, 2013, doi: 10.1016/j.patrec.2013.03.026.
    22. A. Kumar and S. Srivastava, “Restoration and enhancement of breast ultrasound images using extended complex diffusion based unsharp masking,” Proc. Inst. Mech. Eng. Part H J. Eng. Med., p. 09544119211039317, 2021.
    23. G. Gilboa, N. Sochen, and Y. Y. Zeevi, “Image enhancement and denoising by complex diffusion processes,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 26, no. 8, pp. 1020–1036, 2004, doi: 10.1109/TPAMI.2004.47.
    24. Y. F. Pu, “Fractional-Order Euler-Lagrange Equation for Fractional-Order Variational Method: A Necessary Condition for Fractional-Order Fixed Boundary Optimization Problems in Signal Processing and Image Processing,” IEEE Access, vol. 4, pp. 10110–10135, 2016, doi: 10.1109/ACCESS.2016.2636159.
    25. J. Bergstra and Y. Bengio, “Random search for hyper-parameter optimization,” J. Mach. Learn. Res., vol. 13, pp. 281–305, 2012.
    26. A. Dutta and A. Zisserman, “The {VIA} Annotation Software for Images, Audio and Video,” 2019, doi: 10.1145/3343031.3350535.
    27. X. Xia and B. Kulis, “W-Net: A Deep Model for Fully Unsupervised Image Segmentation,” 2017, [Online]. Available: http://arxiv.org/abs/1711.08506.
    28. O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” 2015, pp. 234–241.
    29. D. Misra, “Mish: A Self Regularized Non-Monotonic Activation Function,” 31st Br. Mach. Vis. Conf. BMVC 2020, 2020, doi: 10.5244/c.34.191.
    30. A. Kumar, P. Kumar, and S. Srivastava, “A skewness reformed complex diffusion based unsharp masking for the restoration and enhancement of Poisson noise corrupted mammograms,” Biomed. Signal Process. Control, vol. 73, no. August 2021, p. 103421, 2022, doi: 10.1016/j.bspc.2021.103421.
    31. C. Shorten and T. M. Khoshgoftaar, “A survey on Image Data Augmentation for Deep Learning,” J. Big Data, vol. 6, no. 1, 2019, doi: 10.1186/s40537-019-0197-0.
    32. P. Kumar, A. Kumar, S. Srivastava, and Y. Padma Sai, “A novel bi-modal extended Huber loss function based refined mask RCNN approach for automatic multi instance detection and localization of breast cancer,” Proc. Inst. Mech. Eng. Part H J. Eng. Med., p. 09544119221095416.
    33. K. R. Singh, A. Sharma, and G. K. Singh, “W-Net: Novel Deep Supervision for Deep Learning-based Cardiac Magnetic Resonance Imaging Seg-mentation,” IETE J. Res., vol. 69, no. 12, pp. 8960–8976, 2023, doi: 10.1080/03772063.2022.2098836.

  • Downloads

  • How to Cite

    Nath, M., & Srivastava, S. (2025). MW-Net: A Deep Framework for Segmentation and Detection of S1 and S2 Heart Sounds. International Journal of Basic and Applied Sciences, 14(7), 1-10. https://doi.org/10.14419/k0nqfx70