Abstract
Heart sound signal analysis is an important area in healthcare, and the detection of imbalanced heart sounds can provide valuable diagnostic information. However, due to heart sound variability, accurate prediction of imbalanced signals remains challenging. The issue of class inequality has received much attention from numerous scientific domains. The correct classification becomes increasingly challenging as data scale and data imbalance increase. Traditional classifiers tend to favor the dominant class and overlook the minority class, which is frequently considerably more significant when dealing with imbalanced learning problems. We propose an ensemble learning algorithm based on a transfer learning convolutional neural network (CNN) model to solve these challenges to predict imbalanced heart sound signals. We employ spectrogram images and STFT to extract the relevant features from Phonocardiogram (PCG) data. Our model leverages the pre-trained CNN architecture and fine-tunes it on the spectrogram images to improve the prediction performance. Moreover, we incorporate an ensemble approach to improve the model’s robustness and accuracy. Our experimental results on a publicly available PhysioNet PCG dataset demonstrate that the proposed algorithm outperforms existing state-of-the-art methods in terms of accuracy, sensitivity, and specificity. The ensemble methodology comprising AlexNet, SqueezeNet, and VGG19 models was proposed and achieved the highest level of performance, resulting in an accuracy of 99.20% and a sensitivity rate of 99.47%. Our study showcases the potential of leveraging technological advancements to predict unbalanced Phonocardiogram (PCG) signals using spectrogram images. This research opens up promising avenues for future exploration in cardiac diagnostics. Specifically, the ensemble-based transfer learning model proposed in this study holds great promise.
Similar content being viewed by others
Data Availability
The datasets utilized and/or examined in the current study are publicly accessible through the PhysioNet/CinC Challenge heart sound database [6]. This information is available to other researchers and can potentially contribute to further advancements in the field of heart sound analysis.
References
Classifying heart sounds challenge. https://www.peterjbentley.com/heartchallenge/index.html. Accessed 16 Apr 2023
Ajitkumar Singh S, Dinita Devi N, Majumder S (2022) An improved unsegmented phonocardiogram classification using nonlinear time scattering features. Comput J. https://doi.org/10.1093/comjnl/bxac025
Martinez-Alajarin J, Ruiz-Merino R (2005) Efficient method for events detection in phonocardiographic signals. Bioeng Bioinspired Syst II, vol 5839, p 398. https://doi.org/10.1117/12.608203. October 2014
Hamidi M, Ghassemian H, Imani M (2018) Classification of heart sound signal using curve fitting and fractal dimension. Biomed Signal Process Control 39:351–359. https://doi.org/10.1016/j.bspc.2017.08.002
Zhou X, Liang W, Wang KIK, Wang H, Yang LT, Jin Q (2020) Deep-learning-enhanced human activity recognition for internet of healthcare things. IEEE Internet Things J 7(7):6429–6438. https://doi.org/10.1109/JIOT.2020.2985082
Liu C et al (2016) An open access database for the evaluation of heart sound algorithms. Physiol Meas 37(12):2181. https://doi.org/10.1088/0967-3334/37/12/2181
Yaseen, Son GY, Kwon S (2018) Classification of heart sound signal using multiple features. Appl Sci 8(12). https://doi.org/10.3390/app8122344
Pathak A, Mandana K, Saha G (2022) Ensembled transfer learning and multiple kernel learning for phonocardiogram based atherosclerotic coronary artery disease detection. IEEE J Biomed Heal Inform 26(6):2804–2813. https://doi.org/10.1109/JBHI.2022.3140277
Samanta P, Pathak A, Mandana K, Saha G (2019) Classification of coronary artery diseased and normal subjects using multi-channel phonocardiogram signal. Biocybern Biomed Eng 39(2):426–443. https://doi.org/10.1016/J.BBE.2019.02.003
Li H et al (2020) A fusion framework based on multi-domain features and deep learning features of phonocardiogram for coronary artery disease detection. Comput Biol Med 120:103733. https://doi.org/10.1016/J.COMPBIOMED.2020.103733
Pathak A, Samanta P, Mandana K, Saha G (2020) Detection of coronary artery atherosclerotic disease using novel features from synchrosqueezing transform of phonocardiogram. Biomed Signal Process Control 62:102055. https://doi.org/10.1016/J.BSPC.2020.102055
Singh SA, Majumder S (2019) Classification of unsegmented heart sound recording using knn classifier. J Mech Med Biol 19(4):1950025. https://doi.org/10.1142/S0219519419500258
Mishra M, Banerjee S, Thomas DC, Dutta S, Mukherjee A (2018) Detection of third heart sound using variational mode decomposition. IEEE Trans Instrum Meas 67(7):1713–1721. https://doi.org/10.1109/TIM.2018.2805198
Maity A, Pathak A, Saha G (2023) Transfer learning based heart valve disease classification from Phonocardiogram signal. Biomed Signal Process Control 85:104805. https://doi.org/10.1016/j.bspc.2023.104805. August 2022
Singh SA, Majumder S (2020) Short unsegmented PCG classification based on ensemble classifier. Turkish J Electr Eng Comput Sci 28(2):875–889. https://doi.org/10.3906/elk-1905-165
Babaei S, Geranmayeh A (2009) Heart sound reproduction based on neural network classification of cardiac valve disorders using wavelet transforms of PCG signals. Comput Biol Med 39(1):8–15. https://doi.org/10.1016/J.COMPBIOMED.2008.10.004
Karhade J, Dash S, Ghosh SK, Dash DK, Tripathy RK (2022) Time-Frequency-Domain Deep Learning Framework for the Automated Detection of Heart Valve Disorders Using PCG Signals. IEEE Trans Instrum Meas 71. https://doi.org/10.1109/TIM.2022.3163156
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. IEEE Trans Biomed Circuits Syst 12(1):24–34. https://doi.org/10.1109/TBCAS.2017.2751545
Gupta S, Agrawal M, Deepak D (2021) Gammatonegram based triple classification of lung sounds using deep convolutional neural network with transfer learning. Biomed Signal Process Control 70:102947. https://doi.org/10.1016/j.bspc.2021.102947
Ismail S, Ismail B (2023) PCG signal classification using a hybrid multi round transfer learning classifier. Biocybern Biomed Eng 43(1):313–334
Jamil S, Roy AM (2023) An efficient and robust Phonocardiography (PCG)-based Valvular Heart Diseases (VHD) detection framework using Vision Transformer (ViT). Comput Biol Med 158:106734
Zang J, Lian C, Xu B, Zhang Z, Su Y, Xue C (2023) AmtNet: Attentional multi-scale temporal network for phonocardiogram signal classification. Biomed Signal Process Control 85:104934
Riccio D, Brancati N, Sannino G, Verde L, Frucci M (2023) CNN-based classification of phonocardiograms using fractal techniques. Biomed Signal Process Control 86:105186
Azam FB, Ansari MI, Nuhash SISK, McLane I, Hasan T (2022) Cardiac anomaly detection considering an additive noise and convolutional distortion model of heart sound recordings. Artif Intell Med 133:102417. https://doi.org/10.1016/j.artmed.2022.102417
Thanaraj P, Parvathavarthini K, Snekhalatha B (2020) Automated heart sound classification system from unsegmented phonocardiogram ( PCG ) using deep neural network. Phys Eng Sci Med 43:505–515. https://doi.org/10.1007/s13246-020-00851-w
Er MB (2021) Heart sounds classification using convolutional neural network with 1D-local binary pattern and 1D-local ternary pattern features. Appl Acoust 180:108152. https://doi.org/10.1016/j.apacoust.2021.108152
Kalidas N, Patidar S, Nesaragi N (2021) Automated detection of abnormal heart sound signals using Fano-factor constrained tunable quality wavelet transform. Biocybern Biomed Eng 41:111–126. https://doi.org/10.1016/j.bbe.2020.12.007
Zhang W, Han J, Deng S (2019) Abnormal heart sound detection using temporal quasi-periodic features and long short-term memory without segmentation. Biomed Signal Process Control 53:101560. https://doi.org/10.1016/j.bspc.2019.101560
Xiao B, Xu Y, Bi X, Zhang J, Ma X (2020) Heart sounds classification using a novel 1-D convolutional neural network with extremely low parameter consumption. Neurocomputing 392:153–159. https://doi.org/10.1016/j.neucom.2018.09.101
Chen Y, Wei S (2020) Classification of heart sounds based on the combination of the modified frequency wavelet transform and convolutional neural network. Med Biol Eng Comput 58:2039–2047
Ding H et al (2023) RVGAN-TL: A generative adversarial networks and transfer learning-based hybrid approach for imbalanced data classification. Inf Sci (Ny) 629:184–203. https://doi.org/10.1016/J.INS.2023.01.147
Khan JS, Kaushik M, Chaurasia A, Dutta MK, Burget R (2022) Cardi-Net: A deep neural network for classification of cardiac disease using phonocardiogram signal. Comput Methods Programs Biomed 219:106727. https://doi.org/10.1016/j.cmpb.2022.106727
Wang C, Deng C, Yu Z, Hui D, Gong X, Luo R (2021) Adaptive ensemble of classifiers with regularization for imbalanced data classification. Inf Fusion 69:81–102. https://doi.org/10.1016/J.INFFUS.2020.10.017
Ren J, Wang Y, Mao M, Ming Cheung Y (2022) Equalization ensemble for large scale highly imbalanced data classification. Knowledge-Based Syst 242:108295. https://doi.org/10.1016/J.KNOSYS.2022.108295
Schmidt SE, Toft E, Holst-Hansen C, Graff C, Struijk JJ (2008) Segmentation of heart sound recordings from an electronic stethoscope by a duration dependent Hidden-Markov model. In computers in cardiology vol. 35, pp 345–348. https://doi.org/10.1109/CIC.2008.4749049
Ismail S, Ismail B, Siddiqi I, Akram U (2022) PCG classification through spectrogram using transfer learning. Biomed Signal Process Control 79:104075. no P1. https://doi.org/10.1016/j.bspc.2022.104075
Jablonski A, Dziedziech K (2022) Intelligent spectrogram - A tool for analysis of complex non-stationary signals. Mech Syst Signal Process 167:108554. no PA. https://doi.org/10.1016/j.ymssp.2021.108554
Chen J, Chen J, Zhang D, Sun Y, Nanehkaran YA (2020) Using deep transfer learning for image-based plant disease identification. Comput Electron Agric 173:105393. https://doi.org/10.1016/j.compag.2020.105393
Zhou R et al (2023) An adaptively weighted ensemble of multiple CNNs for carotid ultrasound image segmentation. Biomed Signal Process Control 83:104673. https://doi.org/10.1016/j.bspc.2023.104673
Li R, Gao R, Suganthan PN (2023) A decomposition-based hybrid ensemble CNN framework for driver fatigue recognition. Inf Sci 624:833–848. https://doi.org/10.1016/j.ins.2022.12.088
Stateczny A, Narahari SC, Vurubindi P, Guptha NS, Srinivas K (2023) Underground water level prediction in remote sensing images using improved hydro index value with ensemble classifier. Remote Sens 15(8):2015. https://doi.org/10.3390/rs15082015
Xie Y, Sun W, Ren M, Chen S, Huang Z, Pan X (2023) Stacking ensemble learning models for daily runoff prediction using 1D and 2D CNNs. Expert Syst Appl 217:119469. https://doi.org/10.1016/J.ESWA.2022.119469
Krizhevsky A, Sutskever I, Hinton GE (2017) ImageNet classification with deep convolutional neural networks. Commun ACM 60(6):84–90. https://doi.org/10.1145/3065386
Iandola FN, Han S, Moskewicz MW, Ashraf K, Dally WJ, Keutzer K (2016) SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and ¡0.5MB model size. Feb. [Online]. Available: arXiv:1602.07360v4. Accessed 24 Apr 2023
Simonyan K, Zisserman A (2014) Very Deep Convolutional Networks for Large-Scale Image Recognition. 3rd Int. Conf. Learn. Represent. ICLR 2015 - Conf Track Proc, Sep. [Online]. Available: arXiv:1409.1556v6. Accessed 24 Apr 2023
Hazeri H, Zarjam P, Azemi G (2021) Classification of normal/abnormal PCG recordings using a time-frequency approach. Analog Integr Circuits Signal Process 109(2):459–465. https://doi.org/10.1007/s10470-021-01867-2
Zeng W, Yuan J, Yuan C, Wang Q, Liu F, Wang Y (2020) A new approach for the detection of abnormal heart sound signals using TQWT, VMD and neural networks. Artif Intell Rev 54:1613–1647. https://doi.org/10.1007/s10462-020-09875-w
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Ethics approval
There is no issue with Ethical approval and Informed consent
Conflict of Interests
There is no conflict of interest
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Singh, S.A., Devi, N.D., Singh, K.N. et al. An ensemble-based transfer learning model for predicting the imbalance heart sound signal using spectrogram images. Multimed Tools Appl 83, 39923–39942 (2024). https://doi.org/10.1007/s11042-023-17186-9
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-023-17186-9