Skip to content
2000
image of Heart Sound Classification Using Kernel Partial Least Squares with Easy MKL-derived Kernels

Abstract

Introduction

The automatic classification of heart sound signals offers an economical and convenient approach for early diagnosis of cardiac diseases. By leveraging technological advancements, this method facilitates early detection and management of heart conditions, which is critical for improving patient outcomes.

Methods

To address the challenges in analyzing complex heart sound signals, we introduce a novel method utilizing EasyMKL-enhanced kernel partial least squares (KPLS). This approach begins with transforming segmented cardiac cycles into the time-frequency domain using the short-time Fourier transform (STFT). The STFT representations are then mapped into a high-dimensional feature space using multiple kernel functions derived from Easy MKL, designed to capture and enhance the discriminative nonlinear relationships among various heart sound categories. The extracted features are classified using a Support Vector Machine (SVM) for datasets with balanced samples and an XGBoost classifier for those with imbalanced samples.

Results

The proposed method was evaluated on two publicly available heart sound datasets, the PhysioNet/CinC Challenge 2016 and the Yaseen dataset. On the PhysioNet/CinC Challenge 2016 dataset, our method achieved a sensitivity of 0.9217, a specificity of 0.8950, and an overall score of 0.9084. On the Yaseen dataset, our method achieved an average recall of 0.9933, precision of 0.9930, and F1-score of 0.9930, demonstrating high classification accuracy across different heart sound categories. These results confirm the effectiveness of our approach in extracting discriminative features and improving classification performance.

Discussion

The high performance across two diverse datasets confirms the generalizability and robustness of the proposed method. Notably, the EasyMKL-enhanced KPLS framework captures complex nonlinear patterns while maintaining interpretability—an essential attribute for clinical applications. Compared to traditional approaches, our method significantly improves feature discriminability, as evidenced by ablation studies. While minor misclassifications persist in acoustically similar classes, the model consistently outperforms baselines, highlighting its strong potential for deployment in real-world intelligent auscultation systems.

Conclusion

The experimental results confirm the superiority of our proposed method, demonstrating its potential as a powerful tool for the automatic classification of heart sound signals. This approach not only enhances the accuracy of cardiac disease diagnostics but also offers a robust framework for handling complex and nonlinear characteristics of heart sound data, promising significant contributions to clinical practices and research in cardiology.

© 2025 Bentham Science Publishers
Loading

Article metrics loading...

/content/journals/cbio/10.2174/0115748936366098250718052357
2025-07-30
2025-10-31

  • From This Site

    /content/journals/cbio/10.2174/0115748936366098250718052357
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Nedkoff L. Briffa T. Zemedikun D. Herrington S. Wright F.L. Global trends in atherosclerotic cardiovascular disease. Clin. Ther. 2023 45 11 1087 1091 10.1016/j.clinthera.2023.09.020 37914585
    [Google Scholar]
  2. Roguin A. Rene theophile hyacinthe laënnec (1781-1826): The man behind the stethoscope. Clin. Med. Res. 2006 4 3 230 235 10.3121/cmr.4.3.230 17048358
    [Google Scholar]
  3. Guo Z. Chen J. He T. Wang W. Abbas H. Lv Z. DS-CNN: Dual-stream convolutional neural networksbased heart sound classification for wearable devices. IEEE Trans. Consum. Electron. 2023 69 4 1186 1194 10.1109/TCE.2023.3247901
    [Google Scholar]
  4. Vashishth N Garg D Raj A Design and implementation of remote PCG (phonocardiography) system using audio exchange bus. Int. J. Interact. Des. Manuf. 2023 1 15 10.1007/s12008‑023‑01420‑7
    [Google Scholar]
  5. Tian D. Chen W. Xu D. Xu L. Xu G. Guo Y. Yao Y. A review of traditional Chinese medicine diagnosis using machine learning: Inspection, auscultation-olfaction, inquiry, and palpation. Comput. Biol. Med. 2024 170 108074 10.1016/j.compbiomed.2024.108074 38330826
    [Google Scholar]
  6. Clifford G.D. Liu C. Moody B. Millet J. Schmidt S. Li Q. Silva I. Mark R.G. Recent advances in heart sound analysis. Physiol. Meas. 2017 38 8 E10 E25 10.1088/1361‑6579/aa7ec8 28696334
    [Google Scholar]
  7. Xiang M. Zang J. Wang J. Wang H. Zhou C. Bi R. Zhang Z. Xue C. Research of heart sound classification using two-dimensional features. Biomed. Signal Process. Control 2023 79 104190 10.1016/j.bspc.2022.104190
    [Google Scholar]
  8. Aji N.B. Kurnianingsih K. Masuyama N. CNN-LSTM for heartbeat sound classification. Inter. J. Infor. Visualizat. 2024 8 2 735741 10.62527/joiv.8.2.2115
    [Google Scholar]
  9. Shastri R.K. Shastri A.R. Nitnaware P.P. Padulkar D.M. Hybrid Sneaky algorithm-based deep neural networks for Heart sound classification using phonocardiogram. Network 2024 35 1 1 26 10.1080/0954898X.2023.2270040 38018148
    [Google Scholar]
  10. Wang Z. Qian K. Liu H. Hu B. Schuller B.W. Yamamoto Y. Exploring interpretable representations for heart sound abnormality detection. Biomed. Signal Process. Control 2023 82 104569 10.1016/j.bspc.2023.104569
    [Google Scholar]
  11. Li S. Sun J. Yang H. Pan J. Guo T. Wang W. Interpretable End-to-End heart sound classification. Measurement 2024 237 115113 10.1016/j.measurement.2024.115113
    [Google Scholar]
  12. Kaur I. Ahmad T. Doja M.N. A systematic review of medical expert systems for cardiac arrest prediction. Curr. Bioinform. 2024 19 6 551 570 10.2174/0115748936251658231002043812
    [Google Scholar]
  13. Kong X. Shen C. Tang J. A review of biosequences alignment, matching, and mining based on GPU. Curr. Bioinform. 2025 20 10.2174/0115748936353476241230105816
    [Google Scholar]
  14. Chen J. Guo Z. Xu X. Jeon G. Camacho D. Artificial intelligence for heart sound classification: A review. Expert Syst. 2024 41 4 13535 10.1111/exsy.13535
    [Google Scholar]
  15. Ren Z. Chang Y. Nguyen T.T. Tan Y. Qian K. Schuller B.W. A comprehensive survey on heart sound analysis in the deep learning era. IEEE Comput. Intell. Mag. 2024 19 3 42 57 10.1109/MCI.2024.3401309
    [Google Scholar]
  16. Chen J. Guo Z. Xu X. Zhang L. Teng Y. Chen Y. Woźniak M. Wang W. A robust deep learning framework based on spectrograms for heart sound classification. IEEE/ACM Trans. Comput. Biol. Bioinformatics 2024 21 4 936 947 10.1109/TCBB.2023.3247433 37027654
    [Google Scholar]
  17. Wang J. Zang J. Yao S. Zhang Z. Xue C. Multiclassification for heart sound signals under multiple networks and multi-view feature. Measurement 2024 225 114022 10.1016/j.measurement.2023.114022
    [Google Scholar]
  18. Karan B. Padhi T. Hilbert domain characterizations of wavelet packets for automated heart sound abnormality detection. Biomed. Signal Process. Control 2024 90 105793 10.1016/j.bspc.2023.105793
    [Google Scholar]
  19. Jia W. Wang Y. Chen R. Ye J. Li D. Yin F. Yu J. Chen J. Shu Q. Xu W. Zchsound: Open-source zju paediatric heart sound database with congenital heart disease. IEEE Trans. Biomed. Eng. 2024 71 8 2278 2286 10.1109/TBME.2023.3348800 38194403
    [Google Scholar]
  20. Alkhodari M. Fraiwan L. Convolutional and recurrent neural networks for the detection of valvular heart diseases in phonocardiogram recordings. Comput. Methods Programs Biomed. 2021 200 105940 10.1016/j.cmpb.2021.105940 33494031
    [Google Scholar]
  21. Fan J. Tang S. Duan H. Bi X. Xiao B. Li W. Gao X. Le-lwtnet: A learnable lifting wavelet convolutional neural network for heart sound abnormality detection. IEEE Trans. Instrum. Meas. 2023 72 1 14 10.1109/TIM.2023.3246513
    [Google Scholar]
  22. Zhang H. Zhang P. Wang Z. Multi-feature decision fusion network for heart sound abnormality detection and classification. IEEE J. Biomed. Health Inform. 2023 28 3 1386 1397 10.1109/JBHI.2023.3307870 37610909
    [Google Scholar]
  23. Yaseen Son G-Y. Kwon S. Classification of heart sound signal using multiple features. Appl. Sci. 2018 8 12 2344 10.3390/app8122344
    [Google Scholar]
  24. Zhang X. Liu X. Liu G. A heart sound signal classification method based on the mixed characteristics of mel cepstrum coefficient and second-order spectrum. Circuits Syst. Signal Process. 2024 43 6 3533 3552 10.1007/s00034‑023‑02588‑9
    [Google Scholar]
  25. Sadeghi Nia P. Danandeh Hesar H. Abnormal heart sound detection using time-frequency analysis and machine learning techniques. Biomed. Signal Process. Control 2024 90 105899 10.1016/j.bspc.2023.105899
    [Google Scholar]
  26. Roy T.S. Roy J.K. Mandal N. Improvement in the performance of deep learning based on CNN to classify the heart sound by evaluating hyper-parameters. Soft Comput. 2024 28 19 11587 11600 10.1007/s00500‑024‑09909‑3
    [Google Scholar]
  27. Nguyen M.T. Lin W.W. Huang J.H. Heart sound classification using deep learning techniques based on log-mel spectrogram. Circuits Syst. Signal Process. 2023 42 1 344 360 10.1007/s00034‑022‑02124‑1
    [Google Scholar]
  28. Fernando T. Ghaemmaghami H. Denman S. Sridharan S. Hussain N. Fookes C. Heart sound segmentation using bidirectional LSTMs with attention. IEEE J. Biomed. Health Inform. 2020 24 6 1601 1609 10.1109/JBHI.2019.2949516 31670683
    [Google Scholar]
  29. Wang R. Duan Y. Li Y. Zheng D. Liu X. Lam C.T. Tan T. PCTMF-Net: Heart sound classification with parallel CNNs-transformer and second-order spectral analysis. Vis. Comput. 2023 39 8 3811 3822 10.1007/s00371‑023‑03031‑5
    [Google Scholar]
  30. Fakhry M. Gallardo-Antol’ın A. Variational mode decomposition and a light CNN-LSTM model for classification of heart sound signals. EUROCON International Conference on Smart Technologies 2023, pp. 295-300. 2023
    [Google Scholar]
  31. Megalmani D.R. Shailesh B.G. Rao A. Unsegmented heart sound classification using hybrid CNN-LSTM neural networks. IEEE, 2021, pp. 713-717.
    [Google Scholar]
  32. Xiao F. Liu H. Lu J. A new approach based on a 1D + 2D convolutional neural network and evolving fuzzy system for the diagnosis of cardiovascular disease from heart sound signals. Appl. Acoust. 2024 216 109723 10.1016/j.apacoust.2023.109723
    [Google Scholar]
  33. Guo B. Tang H. A framework of heart sound noise reduction using multi-channel structure. Biomed. Signal Process. Control 2024 95 106345 10.1016/j.bspc.2024.106345
    [Google Scholar]
  34. Hossain A. Uddin S. Rahman P. Anee M.J. Rifat M.M.H. Uddin M.M. Wavelet and spectral analysis of normal and abnormal heart sound for diagnosing cardiac disorders. BioMed Res. Int. 2022 2022 1 9092346 10.1155/2022/9092346 35937404
    [Google Scholar]
  35. Mekahlia M.S. Fezari M. Aliouat A. A comparative analysis of Constant-Q Transform, gammatonegram, and Mel-spectrogram techniques for AI-aided cardiac diagnostics. Med. Eng. Phys. 2025 137 104302 10.1016/j.medengphy.2025.104302 40057368
    [Google Scholar]
  36. Sun S. Segmentation-based adaptive feature extraction combined with mahalanobis distance classification criterion for heart sound diagnostic system. IEEE Sens. J. 2021 21 9 11009 11022 10.1109/JSEN.2021.3063222
    [Google Scholar]
  37. Kui H. Pan J. Zong R. Yang H. Wang W. Heart sound classification based on log Mel-frequency spectral coefficients features and convolutional neural networks. Biomed. Signal Process. Control 2021 69 102893 10.1016/j.bspc.2021.102893
    [Google Scholar]
  38. Sun S. Wang H. Cheng C. PCA-based heart sound feature generation for a ventricular septal defect discrimination. International Computer Conference on Wavelet Active Media Technology and Information Processing (ICCWAMTIP) IEEE, 2017, pp. 128-133.
    [Google Scholar]
  39. Zhang W. Han J. Deng S. Heart sound classification based on scaled spectrogram and partial least squares regression. Biomed. Signal Process. Control 2017 32 20 28 10.1016/j.bspc.2016.10.004
    [Google Scholar]
  40. Wang Y. Zhang X. Ju Y. Liu Q. Zou Q. Zhang Y. Ding Y. Zhang Y. Identification of human microRNA-disease association via low-rank approximation-based link propagation and multiple kernel learning. Front. Comput. Sci. 2024 18 2 182903 10.1007/s11704‑023‑2490‑5
    [Google Scholar]
  41. Shi Q. Zheng K. Li H. Wang B. Liang X. Li X. Wang J. LKLPDA: A low-rank fast kernel learning approach for predicting pirna-disease associations. IEEE/ACM Trans. Comput. Biol. Bioinformatics 2024 21 6 2179 2187 10.1109/TCBB.2024.3452055 39213276
    [Google Scholar]
  42. Wang J. Li Z. Tang C. Liu S. Wan X. Liu X. Multiple kernel clustering with adaptive multi-scale partition selection. IEEE Trans. Knowl. Data Eng. 2024 36 11 6641 6652 10.1109/TKDE.2024.3399738
    [Google Scholar]
  43. Ruiz C. Alaíz C.M. Dorronsoro J.R. A survey on kernel-based multi-task learning. Neurocomputing 2024 577 127255 10.1016/j.neucom.2024.127255
    [Google Scholar]
  44. Shah F. Shah M.S. Akram W. Manzoor A. Mahmoud R.O. Abdelminaam D.S. Sign language recognition using multiple kernel learning: A case study of Pakistan sign language. IEEE Access 2021 9 67548 67558 10.1109/ACCESS.2021.3077386
    [Google Scholar]
  45. Rajesh Kumar T. Suresh G.R. Kalaiselvi K. Optimization-enabled deep convolutional network for the generation of normal speech from non-audible murmur based on multi-kernel-based features. Int. J. Wavelets Multiresolution Inf. Process. 2022 20 4 2250004 10.1142/S0219691322500047
    [Google Scholar]
  46. Rosipal R. Trejo L.J. Kernel partial least squares regression in reproducing kernel hilbert space. J. Mach. Learn. Res. 2001 2 Dec 97 123
    [Google Scholar]
  47. Duma Z.S. Susiluoto J. Lamminpää O. Sihvonen T. Reinikainen S-P. Haario H. Kf-pls: Optimizing kernel partial leastsquares (k-pls) with kernel flows. Chemom. Intell. Lab. Syst. 2024 254 105238 10.1016/j.chemolab.2024.105238
    [Google Scholar]
  48. Allegrini F. Olivieri A.C. Two sides of the same coin: Kernel partial least-squares (KPLS) for linear and non-linear multivariate calibration. A tutorial. Talanta Open 2023 7 100235 10.1016/j.talo.2023.100235
    [Google Scholar]
  49. Luo Z.D. Li H.X. Dynamic partial-least-squares-based fault detection for nonlinear distributed parameter systems. IEEE Trans. Instrum. Meas. 2024 73 1 9 10.1109/TIM.2024.3379078
    [Google Scholar]
  50. Chen Y. Wang D. An improved deep kernel partial least squares and its application to industrial data modeling. IEEE Trans. Industr. Inform. 2024 20 5 7894 7903 10.1109/TII.2024.3359453
    [Google Scholar]
  51. Thant Y M Wakamiya T Nukunudompanich M Kernel regression methods for prediction of materials properties: Recent developments. Chem. Phy. Rev. 2025 6 1 1 8 10.1063/5.0242118
    [Google Scholar]
  52. Huang D.Y. Dong M. Li H. Intelligibility detection of pathological speech using asymmetric sparse kernel partial least squares classifier. International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2014, pp. 3744-3748.
    [Google Scholar]
  53. Aiolli F. Donini M. EasyMKL: A scalable multiple kernel learning algorithm. Neurocomputing 2015 169 215 224 10.1016/j.neucom.2014.11.078
    [Google Scholar]
  54. Pisner D.A. Schnyer D.M. Support vector machine. Machine Learning United States Academic Press 2020 101 121
    [Google Scholar]
  55. Chen T. Guestrin C. A scalable tree boosting system. Proceedings of the Acm Sigkdd International Conference on Knowledge Discovery and Data Mining 2016, pp. 785-794.
    [Google Scholar]
  56. Clifford G.D. Liu C. Moody B. Classification of normal/abnormal heart sound recordings: The physionet/computing in cardiology challenge 2016. Computing in Cardiology Conference (CinC) 2016, pp. 609-612.
    [Google Scholar]
  57. Springer D.B. Tarassenko L. Clifford G.D. Logistic regression-HSMM-based heart sound segmentation. IEEE Trans. Biomed. Eng. 2016 63 4 822 832 26340769
    [Google Scholar]
  58. Cherif L.H. Mostafi M. Debbal S.M. Digitals filters in heart sound analysis. Inter. J. Clin. Med. Res. 2014 1 3 97 108
    [Google Scholar]
  59. Pathak A. Mandana K. Saha G. Ensembled transfer learning and multiple kernel learning for phonocardiogram based atherosclerotic coronary artery disease detection. IEEE J. Biomed. Health Inform. 2022 26 6 2804 2813 10.1109/JBHI.2022.3140277 34982707
    [Google Scholar]
  60. Li J. Zhang W. Heart sound classification based on discriminative dictionary learning and support vector machine with multiple kernel learning. Proceedings of the International Conference on Artificial Intelligence, Big Data and Algorithms 2024.
    [Google Scholar]
  61. Quayson E. Ganaa E.D. Zhu Q. Shen X-J. Multi-view representation induced kernel ensemble support vector machine. Neural Process. Lett. 2023 55 6 7035 7056 10.1007/s11063‑023‑11250‑z
    [Google Scholar]
  62. Zhang Z. Sun H. Li S. He J. Cao J. Cui G. Wang G. Two-stage sparse multi-kernel optimization classifier method for more accurate and explainable prediction. Expert Syst. Appl. 2023 231 120635 10.1016/j.eswa.2023.120635
    [Google Scholar]
  63. Ding X. Cui M. Li Y. Chen S. A maximal accuracy and minimal difference criterion for multiple kernel learning. Expert Syst. Appl. 2024 254 124378 10.1016/j.eswa.2024.124378
    [Google Scholar]
  64. Cai Q. Cui G.C. Wang H.X. EEG-based emotion recognition using multiple kernel learning. Mach. Intellig. Res. 2022 19 5 472 484 10.1007/s11633‑022‑1352‑1
    [Google Scholar]
  65. Zhou T. Peng Y. Guo T. Partial least squares-based polynomial chaos Kriging for high-dimensional reliability analysis. Reliab. Eng. Syst. Saf. 2023 240 109545 10.1016/j.ress.2023.109545
    [Google Scholar]
  66. Chen Y. Deng X. A deep supervised learning framework based on kernel partial least squares for industrial soft sensing. IEEE Trans. Industr. Inform. 2023 19 3 3178 3187 10.1109/TII.2022.3182023
    [Google Scholar]
  67. Yuan X. Fan Y. Zhou C. Wang X. Zhang G. Fault diagnosis of check valve based on KPLS optimal feature selection and kernel extreme learning machine. Coatings 2022 12 9 1320 10.3390/coatings12091320
    [Google Scholar]
  68. Ochoa-Muñoz A.F. Contreras-Reyes J.E. Multiple factor analysis based on Nipals algorithm to solve missing data problems. Algorithms 2023 16 10 457 10.3390/a16100457
    [Google Scholar]
  69. Kong X. Luo J. Feng X. Liu M. A general quality-related nonlinear process monitoring approach based on input–output kernel PLS. IEEE Trans. Instrum. Meas. 2023 72 1 12 10.1109/TIM.2023.3238692
    [Google Scholar]
  70. Lim Y. Sarama¨ki T. Diniz P.S.R. Liu Q. Fast convergence method for scaling window sidelobe magnitude. IEEE Signal Process. Lett. 2021 28 2078 2081 10.1109/LSP.2021.3116517
    [Google Scholar]
  71. Zhong S. Chen D. Xu Q. Chen T. Optimizing the Gaussian kernel function with the formulated kernel target alignment criterion for two-class pattern classification. Pattern Recognit. 2013 46 7 2045 2054 10.1016/j.patcog.2012.12.012
    [Google Scholar]
  72. Li F. Tang H. Shang S. Mathiak K. Cong F. Classification of heart sounds using convolutional neural network. Appl. Sci. 2020 10 11 3956 10.3390/app10113956
    [Google Scholar]
  73. Baydoun M. Safatly L. Ghaziri H. El Hajj A. Analysis of heart sound anomalies using ensemble learning. Biomed. Signal Process. Control 2020 62 102019 10.1016/j.bspc.2020.102019
    [Google Scholar]
  74. Xiao B. Xu Y. Bi X. Zhang J. Ma X. Heart sounds classification using a novel 1-D convolutional neural network with extremely low parameter consumption. Neurocomputing 2020 392 153 159 10.1016/j.neucom.2018.09.101
    [Google Scholar]
  75. Gaona A.J. Arini P.D. Deep recurrent learning for heart sounds segmentation based on instantaneous frequency features. arXiv:2201.11320 2022 1 7 10.48550/arXiv.2201.11320
    [Google Scholar]
  76. Ghosh S.K. Ponnalagu R.N. Tripathy R.K. Acharya U.R. Automated detection of heart valve diseases using chirplet transform and multiclass composite classifier with PCG signals. Comput. Biol. Med. 2020 118 103632 10.1016/j.compbiomed.2020.103632 32174311
    [Google Scholar]
  77. Upretee P. Yu¨ksel M.E. Accurate classification of heart sounds for disease diagnosis by a single time-varying spectral feature: Preliminary results. 2019, pp. 1-4.
    [Google Scholar]
/content/journals/cbio/10.2174/0115748936366098250718052357
Loading
Heart Sound Classification Using Kernel Partial Least Squares with Easy MKL-derived Kernels
Bentham Science Publishers ; https://doi.org/10.2174/0115748936366098250718052357
/content/journals/cbio/10.2174/0115748936366098250718052357
/content/journals/cbio/10.2174/0115748936366098250718052357
Loading

Data & Media loading...


  • Article Type:     Research Article
Keywords: multiple kernel learning ; discriminative features ; automatic diagnosis ; EasyMKL ; kernel partial least squares ; Heart sound classification

Most Cited Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error
Please enter a valid_number test