S1 and S2 Heart Sound Recognition Using Deep Neural Networks
The DOI matches its Crossref record; 35 of the 35 references it cites were found in a registry or an index.
Abstract
OBJECTIVE: This study focuses on the first (S1) and second (S2) heart sound recognition based only on acoustic characteristics; the assumptions of the individual durations of S1 and S2 and time intervals of S1-S2 and S2-S1 are not involved in the recognition process. The main objective is to investigate whether reliable S1 and S2 recognition performance can still be attained under situations where the duration and interval information might not be accessible. METHODS: A deep neural network (DNN) method is proposed for recognizing S1 and S2 heart sounds. In the proposed method, heart sound signals are first converted into a sequence of Mel-frequency cepstral coefficients (MFCCs). The K-means algorithm is applied to cluster MFCC features into two groups to refine their representation and discriminative capability. The refined features are then fed to a DNN classifier to perform S1 and S2 recognition. We conducted experiments using actual heart sound signals recorded using an electronic stethoscope. Precision, recall, F-measure, and accuracy are used as the evaluation metrics. RESULTS: The proposed DNN-based method can achieve high precision, recall, and F-measure scores with more than 91% accuracy rate. CONCLUSION: The DNN classifier provides higher evaluation scores compared with other well-known pattern classification methods. SIGNIFICANCE: The proposed DNN-based method can achieve reliable S1 and S2 recognition performance based on acoustic characteristics without using an ECG reference or incorporating the assumptions of the individual durations of S1 and S2 and time intervals of S1-S2 and S2-S1.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Book Review: Pattern Recognition and Machine Learning
Nasrabadi, N. M. (2007) · Journal of Electronic Imaging · doi:10.1117/1.2819119 · Cited by 22,066 · cited 1×
“Subsequently, the K-means algorithm [15] is used to divide one heart sound fragment into two groups.”
MethodHinton, G., Deng, L., Yu, D., Dahl, G., Mohamed, A.-R., Jaitly, N., Senior, A., Vanhoucke, V., Nguyen, P., Sainath, T., et al. (2012) · IEEE Signal Processing Magazine · doi:10.1109/msp.2012.2205597 · Cited by 10,260
Learning Deep Architectures for AI
Bengio, Y. (2009) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000006 · Cited by 9,239
Harrell, F. E. (2001) · Springer Series in Statistics · doi:10.1007/978-1-4757-3462-1 · Cited by 9,195
Pattern Recognition and Machine Learning
(2007) · Kybernetes · doi:10.1108/03684920710743466 · Cited by 8,265
Fundamentals of speech recognition
Rabiner, L. R., & Juang, B.-H. F. (1993) · Cited by 7,680
(2011) · Cited by 6,993
An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression
Altman, N. S. (1992) · The American Statistician · doi:10.1080/00031305.1992.10475879 · Cited by 5,708 · cited 1×
“To demonstrate the effectiveness of DNN for S1 and S2 recognition, we compare it with otherwell-known classifiers, namelyKnearest neighbor (KNN) [21], Gaussian mixture model (GMM) [22], logistic regression (LR) [23], and support vector machine (SVM) [24].”
MethodDavis, S., & Mermelstein, P. (1980) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/tassp.1980.1163420 · Cited by 5,465 · cited 1×
“MFCC feature extraction has been confirmed to be effective in speech recognition [25], speaker”
Background(2010) · Journal of Machine Learning Research · doi:10.5555/1756006.1953039 · Cited by 4,051 · cited 1×
“‖F denotes the Frobenius norm, and ρ(A) is the sparsity penalty of the hidden outputs [35], [36]; γ and η are the controlling coefficients.”
BackgroundMulti-column deep neural networks for image classification
Ciresan, D., Meier, U., & Schmidhuber, J. (2012) · 2012 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2012.6248110 · Cited by 3,720
Context-dependent pre-trained deep neural networks for large-vocabulary speech recognition
Dahl, G. E., Yu, D., Deng, L., & Acero, A. (2012) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/tasl.2011.2134090 · Cited by 3,011
A Practical Guide to Training Restricted Boltzmann Machines
Hinton, G. E. (2012) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-35289-8_32 · Cited by 2,852
Best practices for convolutional neural networks applied to visual document analysis
Simard, P. Y., Steinkraus, D., & Platt, J. C. (2005) · Seventh International Conference on Document Analysis and Recognition, 2003. Proceedings. · doi:10.1109/icdar.2003.1227801 · Cited by 2,741 · cited 1×
“More recently, NNs with multiple layers (known as DNN) have shown outstanding performance in a wide variety of tasks, such as automatic speech recognition [18], speech processing [29], [30], and visual pattern recognition [31], [32].”
BackgroundLearning and development in neural networks: the importance of starting small
Elman, J. L. (1993) · Cognition · doi:10.1016/0010-0277(93)90058-4 · Cited by 1,988 · cited 1×
“Finally, the supervector is fed into a deep NN (DNN) classifier [16]–[20] in order to classify S1 and S2.”
MethodRao, S. J. (2003) · Journal of the American Statistical Association · doi:10.1198/jasa.2003.s263 · Cited by 1,941
Acoustic Modeling Using Deep Belief Networks
Mohamed, A.-R., Dahl, G. E., & Hinton, G. (2012) · IEEE Transactions on Audio, Speech, and Language Processing · doi:10.1109/tasl.2011.2109382 · Cited by 1,727 · cited 5×
“The contrastive divergence algorithm is generally adopted for estimating the parameters in RBMs [20].”
MethodInfluentialSpeech Analysis Synthesis and Perception
Flanagan, J. L. (1972) · doi:10.1007/978-3-662-01562-9 · Cited by 1,469
The DET curve in assessment of detection task performance
Martin, A., Doddington, G., Kamm, T., Ordowski, M., & Przybocki, M. (1997) · 5th European Conference on Speech Communication and Technology (Eurospeech 1997) · doi:10.21437/eurospeech.1997-504 · Cited by 1,384 · cited 1×
“These metrics serve as the standards for evaluating the effectiveness of a pattern recognition system [38].”
BackgroundA Regression Approach to Speech Enhancement Based on Deep Neural Networks
Xu, Y., Du, J., Dai, L.-R., & Lee, C.-H. (2015) · IEEE/ACM Transactions on Audio, Speech, and Language Processing · doi:10.1109/taslp.2014.2364452 · Cited by 1,366 · cited 1×
“More recently, NNs with multiple layers (known as DNN) have shown outstanding performance in a wide variety of tasks, such as automatic speech recognition [18], speech processing [29], [30], and visual pattern recognition [31], [32].”
BackgroundCepstral analysis technique for automatic speaker verification
Furui, S. (1981) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/tassp.1981.1163530 · Cited by 1,173 · cited 2×
“Hence, three times of dimensions of original features can be obtained after appending velocity and acceleration features [26].”
BackgroundExploring Strategies for Training Deep Neural Networks
Larochelle, H., Bengio, Y., Louradour, J., & Lamblin, P. (2009) · doi:10.1145/1577069.1577070 · Cited by 1,095
Mel Frequency Cepstral Coefficients For Music Modeling.
Logan, B. (2000) · doi:10.5281/zenodo.1416443 · Cited by 913
Speech enhancement based on deep denoising autoencoder
Lu, X., Tsao, Y., Matsuda, S., & Hori, C. (2013) · Interspeech 2013 · doi:10.21437/interspeech.2013-130 · Cited by 834 · cited 1×
“More recently, NNs with multiple layers (known as DNN) have shown outstanding performance in a wide variety of tasks, such as automatic speech recognition [18], speech processing [29], [30], and visual pattern recognition [31], [32].”
BackgroundAutomatic Speech Recognition: A Deep Learning Approach
Yu, D., & Deng, L. (2015) · Signals and Communication Technology · doi:10.1007/978-1-4471-5779-3 · Cited by 474 · cited 1×
“In this study, cross-entropy [34] is used as the cost function.”
MethodHeart sound segmentation algorithm based on heart sound envelogram
Liang, H., Lukkarinen, S., & Hartimo, I. (2002) · Computers in Cardiology 1997 · doi:10.1109/cic.1997.647841 · Cited by 414 · cited 1×
“1 and based on Shannon energy [43] was applied to detect heart sound segments before performing classification.”
MethodLearning with Support Vector Machines
Campbell, C., & Ying, Y. (2011) · Synthesis lectures on artificial intelligence and machine learning · doi:10.2200/s00324ed1v01y201102aim010 · Cited by 226
Feature Learning in Deep Neural Networks - Studies on Speech Recognition Tasks
Yu, D., Seltzer, M. L., Li, J., Huang, J.-T., & Seide, F. T. B. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1301.3605 · Cited by 174 · cited 1×
“Unlike the MFCC features, we used 24-dimensional Fbank features, which are popularly used in DNN-based ASR systems [40]–[42].”
MethodClassification of heart sounds using an artificial neural network
Ölmez, T., & Dokur, Z. (2002) · Pattern Recognition Letters · doi:10.1016/s0167-8655(02)00281-7 · Cited by 146 · cited 2×
“classification performance [12], [13].”
Computer-Based Detection and Analysis of Heart Sound and Murmur
El‐Segaier, M., Lilja, O., Lukkarinen, S., Sörnmo, L., SEPPONEN, R. E., & Pesonen, E. J. (2005) · Annals of Biomedical Engineering · doi:10.1007/s10439-005-4053-3 · Cited by 133
Detection of S1 and S2 Heart Sounds by High Frequency Signatures
Kumar, D., Carvalho, P., Antunes, M., Henriques, J. M. O., Eugénio, L., Schmidt, R., & Habetha, J. (2006) · Conference proceedings · doi:10.1109/iembs.2006.260735 · Cited by 111 · cited 3×
“In [9], a high frequency marker was developed by applying the fast wavelet transform on heart sound signals to extract the high frequency signatures of the valve closing sounds.”
MethodInfluentialAn artificial neural network approach to automatic speech processing
Siniscalchi, S. M., Svendsen, T. K., & Lee, C. (2014) · Neurocomputing · doi:10.1016/j.neucom.2014.03.005 · Cited by 84 · cited 1×
“To overcome this limitation, a pretraining technique that uses unlabeled data is generally adopted [20], [37].”
MethodBeyond auscultation: Acoustic cardiography in clinical practice
Wen, Y.-N., Lee, A. P., Fang, F., Jin, C., & Yu, C. (2014) · International Journal of Cardiology · doi:10.1016/j.ijcard.2013.12.298 · Cited by 70 · cited 1×
“This task can effectively help doctors diagnose the occurrence of diseases [3].”
BackgroundA New Algorithm for Detection of S1 and S2 Heart Sounds
Kumar, D., Carvalho, P., Antunes, M., Gil, P. S., Henriques, J. M. O., & Eugénio, L. (2006) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2006.1660559 · Cited by 54 · cited 2×
“classification performance [12], [13].”
Speech Recognition Using Long-Span Temporal Patterns in a Deep Network Model
Siniscalchi, S. M., Yu, D., Deng, L., & Lee, C. (2013) · IEEE Signal Processing Letters · doi:10.1109/lsp.2013.2237901 · Cited by 32 · cited 1×
“Unlike the MFCC features, we used 24-dimensional Fbank features, which are popularly used in DNN-based ASR systems [40]–[42].”
Method
Cited by
Deep Learning on 1-D Biosignals: a Taxonomy-based Survey
Ganapathy, N., Swaminathan, R., & Deserno, T. (2018) · Yearbook of Medical Informatics · doi:10.1055/s-0038-1667083 · Cited by 83
Studies on biomedical signal processing: a review
Jalagam, M. K., & Mittal, V. K. (2024) · AIP Conference Proceedings · doi:10.1063/5.0159010 · Cited by 4
Metadata
- Venue
- IEEE Transactions on Biomedical Engineering
- Published
- 2017-02
- Volume
- 64
- Issue
- 2
- Pages
- 372-380
- ISSN
- 0018-9294 (ISSN-L)
- 1558-2531
- OpenAlex
- W2342504937
- Cited by
- 202
- References
- 35
- FWCI
- 16.42
- Type
- journal article
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.