Classification of Heart Sounds Using Convolutional Neural Network

Heart sounds play an important role in the diagnosis of cardiac conditions. Due to the low signal-to-noise ratio (SNR), it is problematic and time-consuming for experts to discriminate different kinds of heart sounds. Thus, objective classification of heart sounds is essential. In this study, we combined a conventional feature engineering method with deep learning algorithms to automatically classify normal and abnormal heart sounds. First, 497 features were extracted from eight domains. Then, we fed these features into the designed convolutional neural network (CNN), in which the fully connected layers that are usually used before the classification layer were replaced with a global average pooling layer to obtain global information about the feature maps and avoid overfitting. Considering the class imbalance, the class weights were set in the loss function during the training process to improve the classification algorithm’s performance. Stratified five-fold cross-validation was used to evaluate the performance of the proposed method. The mean accuracy, sensitivity, specificity and Matthews correlation coefficient observed on the PhysioNet/CinC Challenge 2016 dataset were 86.8%, 87%, 86.6% and 72.1% respectively. The proposed algorithm’s performance achieves an appropriate trade-off between sensitivity and specificity.

[1]  Mario Spagnuolo,et al.  Computer analysis of phonocardiograms , 1963 .

[2]  J. Gray,et al.  Helping family physicians improve their cardiac auscultation skills with an interactive CD‐ROM , 2002, The Journal of continuing education in the health professions.

[3]  A. Merla,et al.  Photoplethysmographic Prediction of the Ankle-Brachial Pressure Index through a Machine Learning Approach , 2020, Applied Sciences.

[4]  Yan Liu,et al.  Normal / abnormal heart sound recordings classification using convolutional neural network , 2016, 2016 Computing in Cardiology Conference (CinC).

[5]  Lorenzo Rosasco,et al.  Theory of Deep Learning III: explaining the non-overfitting puzzle , 2017, ArXiv.

[6]  D. Chicco,et al.  The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation , 2020, BMC Genomics.

[7]  J. Suykens,et al.  Indefinite kernels in least squares support vector machines and principal component analysis , 2017 .

[8]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[9]  Cesare Furlanello,et al.  Integrating deep and radiomics features in cancer bioimaging , 2019 .

[10]  Yongwan Park,et al.  Best subsequence selection of heart sound recording based on degree of sound periodicity , 2011 .

[11]  Bryan R. Conroy,et al.  Ensemble of feature-based and deep learning-based classifiers for detection of abnormal heart sounds , 2016, 2016 Computing in Cardiology Conference (CinC).

[12]  A A Luisada,et al.  First heart sound amplitude in experimentally induced alternans. , 1966, Diseases of the chest.

[13]  Huaming Chen,et al.  Discrimination of Aortic and Pulmonary Components from the Second Heart Sound Using Respiratory Modulation and Measurement of Respiratory Split , 2017 .

[14]  Lionel Tarassenko,et al.  Logistic Regression-HSMM-Based Heart Sound Segmentation , 2016, IEEE Transactions on Biomedical Engineering.

[15]  Young-Seob Jeong,et al.  Sentiment Classification Using Convolutional Neural Networks , 2019, Applied Sciences.

[16]  Kumar Sricharan,et al.  Recognizing Abnormal Heart Sounds Using Deep Learning , 2017, KHD@IJCAI.

[17]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[18]  M. T. Villarreal-Molina,et al.  Use of Human Umbilical Vein Endothelial Cells (HUVEC) as a Model to Study Cardiovascular Disease: A Review , 2020, Applied Sciences.

[19]  Özal Yildirim,et al.  A novel wavelet sequence based on deep bidirectional LSTM network model for ECG signal classification , 2018, Comput. Biol. Medicine.

[20]  Gabriel Jimenez-Moreno,et al.  Deep Neural Networks for the Recognition and Classification of Heart Murmurs Using Neuromorphic Auditory Sensors , 2017, IEEE Transactions on Biomedical Circuits and Systems.

[21]  J. Morris,et al.  On the mechanism of production of the heart sounds , 1958 .

[22]  Hong Tang,et al.  Classification of normal/abnormal heart sound recordings based on multi-domain features and back propagation neural network , 2016, 2016 Computing in Cardiology Conference (CinC).

[23]  Fengyu Cong,et al.  Automatic Zebrafish Egg Phenotype Recognition from Bright-Field Microscopic Images Using Deep Convolutional Neural Network , 2019, Applied Sciences.

[24]  Gerald Penn,et al.  Convolutional Neural Networks for Speech Recognition , 2014, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[25]  Algirdas Maknickas,et al.  Recognition of normal–abnormal phonocardiographic signals using deep convolutional neural networks and mel-frequency spectral coefficients , 2017, Physiological measurement.

[26]  Ting Li,et al.  PCG Classification Using Multidomain Features and SVM Classifier , 2018, BioMed research international.

[27]  Laura E. Boucheron,et al.  Low Bit-Rate Speech Coding Through Quantization of Mel-Frequency Cepstral Coefficients , 2012, IEEE Transactions on Audio, Speech, and Language Processing.

[28]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[29]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[30]  D. Beymer,et al.  Finding disease similarity by combining ECG with heart auscultation sound , 2007, 2007 Computers in Cardiology.

[31]  Qiang Chen,et al.  Network In Network , 2013, ICLR.

[32]  J J Struijk,et al.  Segmentation of heart sound recordings by a duration-dependent hidden Markov model , 2010, Physiological measurement.

[33]  Bo Wang,et al.  Deep Learning for Object Saliency Detection and Image Segmentation , 2015, ArXiv.

[34]  Yoav Freund,et al.  Experiments with a New Boosting Algorithm , 1996, ICML.

[35]  Qiao Li,et al.  An open access database for the evaluation of heart sound algorithms , 2016, Physiological measurement.

[36]  Andrew Lowe,et al.  Fundamental Heart Sound Classification using the Continuous Wavelet Transform and Convolutional Neural Networks , 2018, 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[37]  Soonil Kwon,et al.  Classification of Heart Sound Signal Using Multiple Features , 2018, Applied Sciences.

[38]  Claudio Moraga,et al.  The Influence of the Sigmoid Function Parameters on the Speed of Backpropagation Learning , 1995, IWANN.

[39]  Cesare Furlanello,et al.  A Comparison of MCC and CEN Error Measures in Multi-Class Prediction , 2010, PloS one.

[40]  Rajib Rana,et al.  Phonocardiographic Sensing Using Deep Learning for Abnormal Heartbeat Detection , 2018, IEEE Sensors Journal.

[41]  R. Kusukawa,et al.  Hemodynamic Determinants of the Amplitude of the First Heart Sound , 1965, Circulation research.

[42]  Gian Marti,et al.  Heart sound classification using deep structured features , 2016, 2016 Computing in Cardiology Conference (CinC).