TY - GEN
T1 - Learning Optimal Time-Frequency Representations for Heart Sound
T2 - 9th Conference on Sound and Music Technology, CSMT 2021
AU - Wang, Zhihua
AU - Bao, Zhihao
AU - Qian, Kun
AU - Hu, Bin
AU - Schuller, Björn W.
AU - Yamamoto, Yoshiharu
N1 - Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
PY - 2023
Y1 - 2023
N2 - Computer audition based methods have increasingly attracted efforts among the community of digital health. In particular, heart sound analysis can provide a non-invasive, real-time, and convenient (anywhere and anytime) solution for preliminary diagnosis and/or long-term monitoring of patients who are suffering from cardiovascular diseases. Nevertheless, extracting excellent time-frequency features from the heart sound is not an easy task. On the one hand, heart sound belongs to audio signals, which may be suitable to be analysed by classic audio/speech techniques. On the other hand, this kind of sound generated by our human body should contain some characteristics of physiological signals. To this end, we propose a comprehensive investigation on time-frequency methods for analysing the heart sound, i.e., short-time Fourier transformation, wavelet transformation, Hilbert-Huang transformation, and Log-Mel transformation. The time-frequency representations will be automatically learnt via pre-trained deep convolutional neural networks. Experimental results show that all the investigated methods can reach a mean accuracy higher than 60.0%. Moreover, we find that wavelet transformation can beat other methods by reaching the highest mean accuracy of 75.1% in recognising normal or abnormal heart sounds.
AB - Computer audition based methods have increasingly attracted efforts among the community of digital health. In particular, heart sound analysis can provide a non-invasive, real-time, and convenient (anywhere and anytime) solution for preliminary diagnosis and/or long-term monitoring of patients who are suffering from cardiovascular diseases. Nevertheless, extracting excellent time-frequency features from the heart sound is not an easy task. On the one hand, heart sound belongs to audio signals, which may be suitable to be analysed by classic audio/speech techniques. On the other hand, this kind of sound generated by our human body should contain some characteristics of physiological signals. To this end, we propose a comprehensive investigation on time-frequency methods for analysing the heart sound, i.e., short-time Fourier transformation, wavelet transformation, Hilbert-Huang transformation, and Log-Mel transformation. The time-frequency representations will be automatically learnt via pre-trained deep convolutional neural networks. Experimental results show that all the investigated methods can reach a mean accuracy higher than 60.0%. Moreover, we find that wavelet transformation can beat other methods by reaching the highest mean accuracy of 75.1% in recognising normal or abnormal heart sounds.
KW - Computer audition
KW - Deep learning
KW - Digital health
KW - Heart sound
KW - Time-frequency analysis
KW - Transfer learning
UR - http://www.scopus.com/inward/record.url?scp=85137987305&partnerID=8YFLogxK
U2 - 10.1007/978-981-19-4703-2_8
DO - 10.1007/978-981-19-4703-2_8
M3 - Conference contribution
AN - SCOPUS:85137987305
SN - 9789811947025
T3 - Lecture Notes in Electrical Engineering
SP - 93
EP - 104
BT - Proceedings of the 9th Conference on Sound and Music Technology - Revised Selected Papers from CMST
A2 - Shao, Xi
A2 - Qian, Kun
A2 - Wang, Xin
A2 - Zhang, Kejun
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 1 June 2022
ER -
