No Access

A FUSION FRAMEWORK BASED ON CEPSTRAL DOMAIN FEATURES FROM PHONOCARDIOGRAM TO PREDICT HEART HEALTH STATUS

SHRUTI ARORA

https://orcid.org/0000-0001-7108-2435

Computer Science & Engineering Department, Thapar Institute of Engineering & Technology, Patiala, Punjab, India

E-mail Address: shruti.arora@thapar.edu

Corresponding author.

Search for more papers by this author

SUSHMA JAIN

Computer Science & Engineering Department, Thapar Institute of Engineering & Technology, Patiala, Punjab, India

E-mail Address: sjain@thapar.edu

Search for more papers by this author

, and

INDERVEER CHANA

Computer Science & Engineering Department, Thapar Institute of Engineering & Technology, Patiala, Punjab, India

E-mail Address: inderveer@thapar.edu

Search for more papers by this author

https://doi.org/10.1142/S0219519421500342Cited by:1 (Source: Crossref)

Abstract

A great increase in the number of cardiovascular cases has been a cause of serious concern for the medical experts all over the world today. In order to achieve valuable risk stratification for patients, early prediction of heart health can benefit specialists to make effective decisions. Heart sound signals help to know about the condition of heart of a patient. Motivated by the success of cepstral features in speech signal classification, authors have used here three different cepstral features, viz. Mel-frequency cepstral coefficients (MFCCs), gammatone frequency cepstral coefficients (GFCCs), and Mel-spectrogram for classifying phonocardiogram into normal and abnormal. Existing research has explored only MFCCs and Mel-feature set extensively for classifying the phonocardiogram. However, in this work, the authors have used a fusion of GFCCs with MFCCs and Mel-spectrogram, and achieved a better accuracy score of 0.96 with sensitivity and specificity scores as 0.91 and 0.98, respectively. The proposed model has been validated on the publicly available benchmark dataset PhysioNet 2016.

Keywords:

References

1. Rangayyan RM, Lehner RJ , Phonocardiogram signal analysis: A review, Crit Rev Biomed Eng 15(1) :211–236, 1987. Google Scholar
2. Kumar D, Automatic heart sound analysis for cardiovascular disease assessment, PhD thesis, 2015. https://estudogeral.uc.pt/handle/10316/27015. Google Scholar
3. Liu C, Springer D, Li Q, Moody B, Juan RA, Chorro FJ, Castells F, Roig JM, Silva I, Johnson AE , An open access database for the evaluation of heart sound algorithms, Physiol Meas 37(12) :2181, 2016, https://doi.org/10.1088/0967-3334/37/12/2181. Crossref, Web of Science, Google Scholar
4. Jain PK, Tiwari AK , Heart monitoring systems — A review, Comput Biol Med 54 :1–13, 2014, https://doi.org/10.1016/j.compbiomed.2014.08.014. Crossref, Web of Science, Google Scholar
5. Chakrabarti T, Saha S, Roy S, Chel I , Phonocardiogram signal analysis-practices, trends and challenges: A critical review, Proc IEEE Int Conf Workshop on Computing and Communication (IEMCON), Vancouver, BC, Canada, pp. 1–4, 15–17 October 2015, https://doi.org/10.1109/IEMCON.2015.7344426. Google Scholar
6. El-Dahshan ESA, Bassiouni MM , Intelligent methodologies for cardiac sound signals analysis and characterization in cepstrum and time-scale domains, Comput Intell 36(2) :427–458, 2019, https://doi.org/10.1111/coin.12244. Crossref, Web of Science, Google Scholar
7. Dwivedi AK, Imtiaz SA, Rodriguez-Villegas E , Algorithms for automatic analysis and classification of heart sounds–a systematic review, IEEE Access 7 :8316–8345, 2019, https://doi.org/10.1109/access.2018.2889437. Crossref, Web of Science, Google Scholar
8. Sharma G, Umapathy K, Krishnan S , Trends in audio signal feature extraction methods, Appl Acoust 158 :107020, 2020, https://doi.org/10.1016/j.apacoust.2019.107020. Crossref, Web of Science, Google Scholar
9. Krishnan S, Athavale Y , Trends in biomedical signal feature extraction, Biomed Signal Process Control 43 :41–63, 2018, https://doi.org/10.1016/j.bspc.2018.02.008. Crossref, Web of Science, Google Scholar
10. Son GY, Kwon S , Classification of heart sound signal using multiple features, Appl Sci 8(12) :2344, 2018, https://doi.org/10.3390/app8122344. Crossref, Google Scholar
11. Wang H, Zhang C , The application of gammatone frequency cepstral coefficients for forensic voice comparison under noisy conditions, Aust J Forensic Sci 52(5) :1–16, 2019, https://doi.org/10.1080/00450618.2019.1584830. Web of Science, Google Scholar
12. Davis S, Mermelstein P , Comparison of parametric representations for monosyllabic word recognition in continuously spoken sentences, IEEE Trans Acoust Speech Signal Process 28(4) :357–366, 1980, https://doi.org/10.1109/tassp.1980.1163420. Crossref, Google Scholar
13. Valero X, Alias F , Gammatone cepstral coefficients: Biologically inspired features for non-speech audio classification, IEEE Trans Multimed 14(6) :1684–1689, 2012, https://doi.org/10.1109/tmm.2012.2199972. Crossref, Web of Science, Google Scholar
14. Grzeszick R, Plinge A, Fink GA , Bag-of-features methods for acoustic event detection and classification, IEEE/ACM Trans Audio Speech Lang Process 25(6) :1242–1252, 2017, https://doi.org/10.1109/taslp.2017.2690574. Crossref, Google Scholar
15. Deng M, Meng T, Cao J, Wang S, Zhang J, Fan H , Heart sound classification based on improved MFCC features and convolutional recurrent neural networks, Neural Netw 130 :22–32, 2020, https://doi.org/10.1016/j.neunet.2020.06.015. Crossref, Web of Science, Google Scholar
16. Khan FA, Abid A, Khan MS , Automatic heart sound classification from segmented/unsegmented phonocardiogram signals using time and frequency features, Physiol Meas 41(5) :055006, 2020, https://doi.org/10.1088/1361-6579/ab8770. Crossref, Web of Science, Google Scholar
17. Abduh Z, Nehary EA, Wahed MA, Kadah YM , Classification of heart sounds using fractional Fourier transform based Mel-frequency spectral coefficients and traditional classifiers, Biomed Signal Process Control 57 :101788, 2020, https://doi.org/10.1016/j.bspc.2019.101788. Crossref, Web of Science, Google Scholar
18. Ahmad MS, Mir J, Ullah MO, Shahid MLUR, Syed MA , An efficient heart murmur recognition and cardiovascular disorders classification system, Australas Phys Eng Sci Med 42(3) :733–743, 2019, https://doi.org/10.1007/s13246-019-00778-x. Crossref, Web of Science, Google Scholar
19. Hamidi M, Ghassemian H, Imani M , Classification of heart sound signal using curve fitting and fractal dimension, Biomed Signal Process Control 39 :351–359, 2018, https://doi.org/10.1016/j.bspc.2017.08.002. Crossref, Web of Science, Google Scholar
20. Chen TE, Yang SI, Ho LT, Tsai KH, Chen YH, Chang YF, Lai YH, Wang SS, Tsao Y, Wu CC , S1 and S2 heart sound recognition using deep neural networks, IEEE Trans Biomed Eng 64(2) :372–380, 2017, https://doi.org/10.1109/tbme.2016.2559800. Crossref, Web of Science, Google Scholar
21. Abdollahpur M, Ghaffari A, Ghiasi S, Mollakazemi MJ , Detection of pathological heart sounds, Physiol Meas 38(8) :1616, 2017, https://doi.org/10.1088/1361-6579/aa7840. Crossref, Web of Science, Google Scholar
22. Kay E, Agarwal A , DropConnected neural networks trained on time-frequency and inter-beat features for classifying heart sounds, Physiol Meas 38(8) :1645, 2017, https://doi.org/10.1088/1361-6579/aa6a3d. Crossref, Web of Science, Google Scholar
23. Kwak C, Kwon OW , Cardiac disorder classification by heart sound signals using murmur likelihood and hidden Markov model state likelihood, IET Signal Process 6(4) :326–334, 2012, https://doi.org/10.1049/iet-spr.2011.0170. Crossref, Web of Science, Google Scholar
24. Kumaran U, Rammohan SR, Nagarajan SM, Prathik A , Fusion of Mel and gammatone frequency cepstral coefficients for speech emotion recognition using deep C-RNN, Int J Speech Technol 1–12, 2021, https://doi.org/10.1007/s10772-020-09792-x. Web of Science, Google Scholar
25. Dam QB, Nguyen LT, Nguyen ST, Vu NH , Pham C e-breath: Breath detection and monitoring using frequency cepstral feature fusion, Proc IEEE Int Conf Multimedia Analysis and Pattern Recognition (MAPR), Ho Chi Minh City, Vietnam, pp. 1–6, 9–10 May 2019, https://doi.org/10.1109/mapr.2019.8743533. Google Scholar
26. Dua M, Aggarwal RK, Biswas M , GFCC based discriminatively trained noise robust continuous ASR system for Hindi language, J Ambient Intell Humanized Comput 10(6) :2301–2314, 2019, https://doi.org/10.1007/s12652-018-0828-x. Crossref, Web of Science, Google Scholar
27. Liu JM, You M, Li GZ, Wang Z, Xu X, Qiu Z, Xie W, An C, Chen S , Cough signal recognition with gammatone cepstral coefficients, Proc IEEE China Summit and Int Conf Signal and Information Processing, Beijing, China, pp. 160–164, 6–10 July 2013, https://doi.org/10.1109/chinasip.2013.6625319. Google Scholar
28. Ren Z, Qian K, Dong F, Dai Z, Yamamoto Y, Schuller BW, Deep attention-based representation learning for heart sound classification, 2021. arXiv:210104979. Google Scholar
29. Bozkurt B, Germanakis I, Stylianou Y , A study of time-frequency features for CNN-based automatic heart sound classification for pathology detection, Comput Biol Med 100 :132–143, 2018, https://doi.org/10.1016/j.compbiomed.2018.06.026. Crossref, Web of Science, Google Scholar
30. Amlathe P, Standard machine learning techniques in audio beehive monitoring: Classification of audio samples with logistic regression, K-nearest neighbor, random forest and support vector machine, Master of Science (MS) thesis, Utah State University, 2018. Google Scholar
31. Antink CH, Becker J, Leonhardt S, Walter M , Nonnegative matrix factorization and random forest for classification of heart sound recordings in the spectral domain, Proc Computing in Cardiology Conf (CinC), Vancouver, BC, Canada, pp. 809–812, 11–14 September 2016, https://doi.org/10.22489/cinc.2016.235-109. Google Scholar
32. Singh-Miller NE, Singh-Miller N , Using spectral acoustic features to identify abnormal heart sounds, Proc Computing in Cardiology Conf (CinC), Vancouver, BC, Canada, pp. 557–560, 11–14 September 2016, https://doi.org/10.22489/cinc.2016.160-401. Google Scholar
33. Alexander B, Nallathambi G, Selvaraj N , Screening of heart sounds using hidden Markov and gammatone filterbank models, Proc 17th IEEE Int Conf Machine Learning and Applications (ICMLA), Orlando, FL, USA, pp. 1460–1465, 17–20 December 2018, https://doi.org/10.1109/icmla.2018.00237. Google Scholar
34. Alkhawaldeh RS , DGR: Gender recognition of human speech using one-dimensional conventional neural network, Sci Program 2019, https://doi.org/10.1155/2019/7213717. Web of Science, Google Scholar
35. Tak RN, Agrawal DM, Patil HA , Novel phase encoded Mel filterbank energies for environmental sound classification, Shankar B, Ghosh K, Mandal D, Ray S, Zhang D, Pal S (eds.), Pattern Recognition and Machine Intelligence, PReMI 2017, Lecture Notes in Computer Science, Vol. 10597, Springer, Cham, https://doi.org/10.1007/978-3-319-69900-4_40. Google Scholar
36. Arora V, Leekha R, Singh R, Chana I , Heart sound classification using machine learning and phonocardiogram, Mod Phys Lett B 33(26) :1950321, 2019, https://doi.org/10.1142/S0217984919503214. Link, Web of Science, Google Scholar
37. Adiga A, Magimai M, Seelamantula CS , Gammatone wavelet cepstral coefficients for robust speech recognition, 2013 IEEE Int Conf IEEE Region 10 (TENCON 2013), Xi’an, China, pp. 1–4, 22–25 October 2013, https://doi.org/10.1109/tencon.2013.6718948. Google Scholar
38. ESSENTIA, Open-source library and tools for audio and music analysis, description and synthesis. https://essentia.upf.edu/reference/std_GFCC.html. Google Scholar
39. Lexfors L, Johansson M, Audio representation for environmental sound classification using convolutional neural networks, Master’s theses, Mathematical Sciences FMAM05 20182, 2018. https://lup.lub.lu.se/student-papers/search/publication/8964345. Google Scholar
40. Librosa Library. https://librosa.org/doc/main/generated/librosa.feature.melspectrogram.html. Google Scholar
41. Arora V, Ng EYK, Leekha RS, Verma K, Gupta T, Srinivasan K , Health of things model for classifying human heart sound signals using co-occurrence matrix and spectrogram, J Mech Med Biol 20(6) :2050040, 2020, https://doi.org/10.1142/S0219519420500402. Link, Web of Science, Google Scholar