PCG signal classification using a hybrid multi round transfer learning classifier

• Autorzy: Ismail Shahid , Ismail Basit

Identyfikatory

DOI

10.1016/j.bbe.2023.01.004

• Języki publikacji: EN

Abstrakty

EN

Diagnosis of cardiovascular diseases using Phonocardiography(PCG) is a challenging task as signal itself is cyclo-stationary. It has spectral contents which are overlapped by multiple sources having similar spectral contents but acting as noise. Moreover, length variation in the signals and sampling using different equipment also make analysis of these signal a testing task. In this research, authors have introduced a hybrid technique to counter the variations just mentioned. Our technique is composed of high resolution spectrum generation, conversion of spectral contents to Spectrogram and multi round training. Use of fixed length spectral contents makes system independent of signal length. By using Spectrogram, the deep features can be extracted from spectrum which are used as an input to Pre-trained networks (PTNs). Finally, transfer learning is applied with multiple rounds of training. The introduced methodology is validated using multiple datasets having different PCG signals, sampling frequency, signals length and signal quality. From the reported results, it is evident that Chirplet Z transform (CZT) based Spectrogram can be utilized for mutlticlass classification. If CZT based Spectrograms are passed through multi rounds of training, then accuracy can be further increased. The reported results are accurate to 99% in the case of testing for best case scenarios and even in worst case, the results dont fall below 85%. However, an important observation is that they are consistent across the experimental protocols. The computational cost associated with the introduced technique is low which makes it suitable for hardware implementation.

Słowa kluczowe

EN

PCG; phonocardiography; transfer learning; chirp transform; chirplet Z transform; multi round processing; MRP; FFT; GoogleNet

PL

PCG; fonokardiografia; transfer wiedzy; MRP; FFT; sieć Google

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2023
• Tom: Vol. 43, no. 1
• Strony: 313--334
• Opis fizyczny: Bibliogr. 74 poz., rys., tab., wykr.

Twórcy

autor

Ismail Shahid

• Universidad del Atlantico Medio, Spain

autor

Ismail Basit

• Department of Computer Sciences, Allama Iqbal Open University, Islamabad, Pakistan

Bibliografia

• [1] World Health Organization, https://www.who.int/newsroom/fact-sheets/detail/cardiovascular-diseases-(cvds).html. 2019.
• [2] Son G-Y, Kwon S, et al. Classification of heart sound signal using multiple features. Appl Sci 2018;8(12):2344.
• [3] P.C. heart sounds challenge. Peter Bentley et al. 2011, http:// www.peterjbentley.com/heartchallenge/.
• [4] Cochran WT, Cooley JW, Favin DL, Helms HD, Kaenel RA, Lang WW, et al. What is the fast fourier transform? Proc IEEE 1967;55(10):1664-74.
• [5] Brigham EO, Morrow R. The fast fourier transform. IEEE Spectrum 1967;4(12):63-70.
• [6] Hu G-S, Zhu F-F, et al. An improved chirplet transform and its application for harmonics detection. Circ Syst 2011;2 (3):107-11.
• [7] Ghosh SK, Ponnalagu R, Tripathy R, Acharya UR. Automated detection of heart valve diseases using chirplet transform and multiclass composite classifier with pcg signals. Comput Biol Med 2020;118:103632.
• [8] Singh M, Cheema A. Heart sounds classification using feature extraction of phonocardiography signal. Int J Comput Appl 2013;77(4).
• [9] Redlarski G, Gradolewski D, Palkowski A. A system for heart sounds classification. PloS One 2014;9(11):e112673.
• [10] Kotb MA, Nabih H, El Zahraa F, El Falaki M, Shaker CW, Refaey MA, et al. Improving the recognition of heart murmur. Int J Adv Comput Sci Appl 2016;7(7):283-7.
• [11] Naseri H, Homaeinezhad MR, Pourkhajeh H. Noise/spike detection in phonocardiogram signal as a cyclic random process with non-stationary period interval. Comput Biology Med 2013;43(9):1205-13.
• [12] Springer DB, Tarassenko L, Clifford GD. Logistic regressionhsmm-based heart sound segmentation. IEEE Trans Biomed Eng 2015;63(4):822-32.
• [13] Rubin J, Abreu R, Ganguli A, Nelaturi S, Matei I, Sricharan K. Classifying heart sound recordings using deep convolutional neural networks and mel-frequency cepstral coefficients. In: 2016 Computing in cardiology conference (CinC), IEEE; 2016, pp. 813-16.
• [14] Tang H, Chen H, Li T, Zhong M. Classification of normal, abnormal heart sound recordings based on multi-domain features and back propagation neural network. 2016 Computing in Cardiology Conference (CinC). IEEE; 2016. p. 593-6.
• [15] Nogueira DM, Ferreira CA, Jorge AM. Classifying heart sounds using images of mfcc and temporal features. In: EPIA Conference on Artificial Intelligence. Springer; 2017. p. 186-203.
• [16] Ortiz JJG, Phoo CP, Wiens J. Heart sound classification based on temporal alignment techniques. 2016 computing in cardiology conference (CinC). IEEE; 2016. p. 589-92.
• [17] Tseng K-K, Wang C, Huang Y-F, Chen G-R, Yung K-L, Ip W-H. Cross-domain transfer learning for pcg diagnosis algorithm. Biosensors 2021;11(4):127.
• [18] Sawant NK, Patidar S, Nesaragi N, Acharya UR. Automated detection of abnormal heart sound signals using fano-factor constrained tunable quality wavelet transform. Biocybernet Biomed Eng 2021;41(1):111-26.
• [19] Yang W, Xu J, Xiang J, Yan Z, Zhou H, Wen B, et al. Diagnosis of cardiac abnormalities based on phonocardiogram using a novel fuzzy matching feature extraction method. BMC Medical Informat Decis Making 2022;22(1):1-13.
• [20] Altuve M, Suárez L, Ardila J. Fundamental heart sounds analysis using improved complete ensemble emd with adaptive noise. Biocybernet Biomed Eng 2020;40(1): 426-39.
• [21] Lahmiri S, Bekiros S. Complexity measures of high oscillations in phonocardiogram as biomarkers to distinguish between normal heart sound and pathological murmur. Chaos, Solit Fractals 2022;154:111610.
• [22] Zhang A, Wang J, Qu F, He Z. Classification of children’s heart sounds with noise reduction based on variational modal decomposition. Front Med Technol 2022;4.
• [23] Duggento A, Conti A, Guerrisi M, Toschi N. A novel multi-branch architecture for state of the art robust detection of pathological phonocardiograms. Philosoph Trans Roy Soc A 2021;379(2212):20200264.
• [24] Zeng Y, Yang S, Yu X, Lin W, Wang W, Tong J, et al. A multimodal parallel method for left ventricular dysfunction identification based on phonocardiogram and electrocardiogram signals synchronous analysis. Mathe Biosci Eng 2022;19(9):9612-35.
• [25] Samanta P, Pathak A, Mandana K, Saha G. Classification of coronary artery diseased and normal subjects using multichannel phonocardiogram signal. Biocybernet Biomed Eng 2019;39(2):426-43.
• [26] Digital Library, University of North Texas, Denton, Texas, CZT vs FFT: Flexibility vs Speed, https://digital.library.unt. edu/ark:/67531/metadc737484/m2/1/high_res_d/816417.pdf; 2015.
• [27] Chen Y, Wei S, Zhang Y. Classification of heart sounds based on the combination of the modified frequency wavelet transform and convolutional neural network. Med Biol Eng Comput 2020;58(9):2039-47.
• [28] Baccouche A, Garcia-Zapirain B, Castillo Olea C, Elmaghraby A. Ensemble deep learning models for heart disease classification: a case study from mexico. Information 2020;11 (4):207.
• [29] Chen D, Xuan W, Gu Y, Liu F, Chen J, Xia S, et al. Automatic classification of normal-abnormal heart sounds using convolution neural network and long-short term memory. Electronics 2022;11(8):1246.
• [30] Ghosh SK, Ponnalagu R, Tripathy RK, Panda G, Pachori RB. Automated heart sound activity detection from pcg signal using time-frequency-domain deep neural network. IEEE Trans Instrum Meas 2022;71:1-10.
• [31] Krizhevsky A, Sutskever I, Hinton GE. Imagenet classification with deep convolutional neural networks. Adv Neural Informat Process Syst 2012;25.
• [32] Iandola FN, Han S, Moskewicz MW, Ashraf K, Dally WJ, Keutzer K. Squeezenet: Alexnet-level accuracy with 50x fewer parameters and< 0.5 mb model size, arXiv preprint arXiv:1602.07360.
• [33] Singh SA, Majumder S, Mishra M. Classification of short unsegmented heart sound based on deep learning. 2019 IEEE International Instrumentation and Measurement Technology Conference (I2MTC). IEEE; 2019. p. 1-6.
• [34] Boulares M, Alafif T, Barnawi A. Transfer learning benchmark for cardiovascular disease recognition. IEEE Access 2020;8:109475-91.
• [35] Riaz U, Aziz S, Umar Khan M, Zaidi SAA, Ukasha M, Rashid A. A novel embedded system design for the detection and classification of cardiac disorders. Comput Intell 2021;37 (4):1844-64.
• [36] Tian G, Lian C, Xu B, Zang J, Zhang Z, Xue C. Classification of phonocardiogram based on multi-view deep network. Neural Process Lett 2022:1-16.
• [37] Dhar P, Dutta S, Mukherjee V. Cross-wavelet assisted convolution neural network (alexnet) approach for phonocardiogram signals classification. Biomed Signal Process Control 2021;63:102142.
• [38] Tian G, Lian C, Zeng Z, Xu B, Su Y, Zang J, et al. Imbalanced heart sound signal classification based on two-stage trained dsanet. Cognitive Comput 2022:1-14.
• [39] Indu T, Prakash A, Chandran SR, Babu N, Soorya R. Comparison of different machine learning algorithms for cardiac auscultation. 2022 IEEE International Conference on Signal Processing, Informatics, Communication and Energy Systems (SPICES), vol. 1. IEEE; 2022. p. 113-7.
• [40] Khan MU, Aziz S, Iqtidar K, Zaher GF, Alghamdi S, Gull M. A two-stage classification model integrating feature fusion for coronary artery disease detection and classification. Multimedia Tools Appl 2022;81(10):13661-90.
• [41] Hazeri H, Zarjam P, Azemi G. Classification of normal/ abnormal pcg recordings using a time-frequency approach. Analog Integr Circ Sig Process 2021;109(2):459-65.
• [42] Tian G, Lian C, Zeng Z. Integrated res2net combined with seesaw loss for long-tailed pcg signal classification. In: 2021 11th International Conference on Intelligent Control and Information Processing (ICICIP). IEEE; 2021. p. 53-8.
• [43] Rizal A, Adz-Dzikri AA, Fauzi MAG. Classification of normal and abnormal heart sound using continuous wavelet transform and resnet-50.
• [44] Chen T-E, Yang S-I, Ho L-T, Tsai K-H, Chen Y-H, Chang Y-F, et al. S1 and s2 heart sound recognition using deep neural networks. IEEE Trans Biomed Eng 2016;64(2):372-80.
• [45] Potes C, Parvaneh S, Rahman A, Conroy B. Ensemble of feature-based and deep learning-based classifiers for detection of abnormal heart sounds. 2016 computing in cardiology conference (CinC). IEEE; 2016. p. 621-4.
• [46] Khan MU, Aziz S, Iqtidar K, Zaher GF, Alghamdi S, Gull M. A two-stage classification model integrating feature fusion for coronary artery disease detection and classification. Multimedia Tools Appl 2021:1-30.
• [47] Khan SI, Qaisar SM, Pachori RB. Automated classification of valvular heart diseases using fbse-ewt and psr based geometrical features. Biomed Signal Process Control 2022;73:103445.
• [48] Zeng W, Su B, Yuan C, Chen Y. Automatic detection of heart valve disorders using teager-kaiser energy operator, rationaldilation wavelet transform and convolutional neural networks with pcg signals. Artif Intell Rev 2022:1-26.
• [49] Burns J, Ganigara M, Dhar A. Application of intelligent phonocardiography in the detection of congenital heart disease in pediatric patients: A narrative review. Progress Pediatric Cardiol 2022;64:101455.
• [50] Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, et al. Going deeper with convolutions. In: Proceedings of the IEEE conference on computer vision and pattern recognition; 2015, pp. 1-9.
• [51] MATLAB, Pre-trained networks, https:// www.mathworks.com/help/deeplearning/ug/pretrainedconvolutional-neural-networks.html.
• [52] Karhade J, Dash S, Ghosh SK, Dash DK, Tripathy RK. Time- frequency-domain deep learning framework for the automated detection of heart valve disorders using pcg signals. IEEE Trans Instrum Meas 2022;71:1-11.
• [53] Khan KN, Khan FA, Abid A, Olmez T, Dokur Z, Khandakar A, et al. Deep learning based classification of unsegmented phonocardiogram spectrograms leveraging transfer learning. Physiol Measur 2021;42(9):095003.
• [54] Giorgio A, Guaragnella C, Rizzi M. An effective cad system for heart sound abnormality detection. Circ Syst Signal Process 2022:1-26.
• [55] Arslan Ö., Karhan M. Effect of hilbert-huang transform on classification of pcg signals using machine learning. J King Saud Univ-Comput Informat Sci 2022.
• [56] Almanifi ORA, Ab Nasir AF, Razman MAM, Musa RM, Majeed APA. Heartbeat murmurs detection in phonocardiogram recordings via transfer learning. Alexandria Eng J 2022;61 (12):10995-1002.
• [57] Khaled S, Fakhry M, Esmail H, Ezzat A, Hamad E. Analysis of training optimization algorithms in the narx neural network for classification of heart sound signals. Int J Sci Eng Res 2022;13(2):382-90.
• [58] Morshed M, Fattah SA, Saquib M. Automated heart valve disorder detection based on pdf modeling of formant variation pattern in pcg signal. IEEE Access 2022;10:27330-42.
• [59] Wang M, Guo B, Hu Y, Zhao Z, Liu C, Tang H. Transfer learning models for detecting six categories of phonocardiogram recordings. J Cardiovasc Develop Dis 2022;9(3):86.
• [60] Bao X, Xu Y, Kamavuako EN. The effect of signal duration on the classification of heart sounds: A deep learning approach. Sensors 2022;22(6):2261.
• [61] Zubair M. A peak detection algorithm for localization and classification of heart sounds in pcg signals using k-means clustering. 2021.
• [62] Deperlioglu O. Heart sound classification with signal instant energy and stacked autoencoder network. Biomed Signal Process Control 2021;64:102211.
• [63] Nehary E, Abduh Z, Rajan S. A deep convolutional neural network classification of heart sounds using fractional fourier transform. 2021 IEEE International Instrumentation and Measurement Technology Conference (I2MTC). IEEE; 2021. p. 1-5.
• [64] Netto AN, Abraham L. Detection and classification of cardiovascular disease from phonocardiogram using deep learning models. In: 2021 Second International Conference on Electronics and Sustainable Communication Systems (ICESC). IEEE; 2021. p. 1646-51.
• [65] Ranipa K, Zhu W-P, Swamy M. Multimodal cnn fusion architecture with multi-features for heart sound classification. 2021 IEEE International Symposium on Circuits and Systems (ISCAS). IEEE; 2021. p. 1-5.
• [66] Panah DS, Hines A, Mckeever S. Exploring composite dataset biases for heart sound classification. AICS, 2020. p. 145-56.
• [67] Alqudah AM, Alquran H, Qasmieh IA. Classification of heart sound short records using bispectrum analysis approach images and deep learning. Network Model Anal Health Informat Bioinformat 2020;9(1):1-16.
• [68] Shuvo SB, Ali SN, Swapnil SI, Al-Rakhami MS, Gumaei A. Cardioxnet: A novel lightweight deep learning framework for cardiovascular disease classification using heart sound recordings. IEEE Access 2021;9:36955-67.
• [69] Deperlioglu O, Kose U, Gupta D, Khanna A, Sangaiah AK. Diagnosis of heart diseases by a secure internet of health things system based on autoencoder deep neural network. Comput Commun 2020;162:31-50.
• [70] Banerjee R, Ghose A. A semi-supervised approach for identifying abnormal heart sounds using variational autoencoder. In: ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE; 2020. p. 1249-53.
• [71] Baydoun M, Safatly L, Ghaziri H, El Hajj A. Analysis of heart sound anomalies using ensemble learning. Biomed Signal Process Control 2020;62:102019.
• [72] El Badlaoui O, Benba A, Hammouch A. Novel pcg analysis method for discriminating between abnormal and normal heart sounds. Irbm 2020;41(4):223-8.
• [73] Ahmad MS, Mir J, Ullah MO, Shahid MLUR, Syed MA. An efficient heart murmur recognition and cardiovascular disorders classification system. Austral Phys Eng Sci Med 2019;42(3):733-43.
• [74] Sujadevi V, Soman K, Vinayakumar R, Sankar AP. Deep models for phonocardiography (pcg) classification. In: 2017 International Conference on Intelligent Communication and Computational Techniques (ICCT). IEEE; 2017. p. 211-6.

Uwagi

PL

Opracowanie rekordu ze środków MEiN, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2022-2023)