Deep Attention-based Representation Learning for Heart Sound Classification

Cardiovascular diseases are the leading cause of deaths and severely threaten human health in daily life. On the one hand, there have been dramatically increasing demands from both the clinical practice and the smart home application for monitoring the heart status of subjects suffering from chronic cardiovascular diseases. On the other hand, experienced physicians who can perform an efficient auscultation are still lacking in terms of number. Automatic heart sound classification leveraging the power of advanced signal processing and machine learning technologies has shown encouraging results. Nevertheless, human hand-crafted features are expensive and time-consuming. To this end, we propose a novel deep representation learning method with an attention mechanism for heart sound classification. In this paradigm, high-level representations are learnt automatically from the recorded heart sound data. Particularly, a global attention pooling layer improves the performance of the learnt representations by estimating the contribution of each unit in feature maps. The Heart Sounds Shenzhen (HSS) corpus (170 subjects involved) is used to validate the proposed method. Experimental results validate that, our approach can achieve an unweighted average recall of 51.2 % for classifying three categories of heart sounds, i. e., normal, mild, and moderate/severe annotated by cardiologists with the help of Echocardiography.

[1]  Yineng Zheng,et al.  A novel hybrid energy fraction and entropy-based approach for systolic heart murmurs identification , 2015, Expert Syst. Appl..

[2]  Francesca N. Delling,et al.  Heart Disease and Stroke Statistics—2019 Update: A Report From the American Heart Association , 2019, Circulation.

[3]  Feng Xu,et al.  Portable microfluidic and smartphone-based devices for monitoring of cardiovascular diseases at the point of care. , 2016, Biotechnology advances.

[4]  Nadeem Akhtar,et al.  Interpretation of intelligence in CNN-pooling processes: a methodological survey , 2019, Neural Computing and Applications.

[5]  Yoshua Bengio,et al.  Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling , 2014, ArXiv.

[6]  Shabnam Ghaffarzadegan,et al.  An Ensemble of Transfer, Semi-supervised and Supervised Learning Methods for Pathological Heart Sound Classification , 2018, INTERSPEECH.

[7]  Gábor Gosztolya,et al.  General Utterance-Level Feature Extraction for Classifying Crying Sounds, Atypical & Self-Assessed Affect and Heart Beats , 2018, INTERSPEECH.

[8]  Zhiming Luo,et al.  A novel recurrent hybrid network for feature fusion in action recognition , 2017, J. Vis. Commun. Image Represent..

[9]  Björn Schuller,et al.  Computer Audition for Healthcare: Opportunities and Challenges , 2020, Frontiers in Digital Health.

[10]  Syed Anas Imtiaz,et al.  Algorithms for Automatic Analysis and Classification of Heart Sounds–A Systematic Review , 2019, IEEE Access.

[11]  CALDER SHEAGREN,et al.  UNCERTAINTY PRINCIPLES WITH FOURIER ANALYSIS , 2005 .

[12]  Qiao Li,et al.  Recent advances in heart sound analysis , 2017, Physiological measurement.

[13]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[14]  Björn W. Schuller,et al.  Deep Wavelets for Heart Sound Classification , 2019, 2019 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS).

[15]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[16]  Björn W. Schuller,et al.  The INTERSPEECH 2018 Computational Paralinguistics Challenge: Atypical & Self-Assessed Affect, Crying & Heart Beats , 2018, INTERSPEECH.

[17]  Björn W. Schuller,et al.  Machine Listening for Heart Status Monitoring: Introducing and Benchmarking HSS—The Heart Sounds Shenzhen Corpus , 2019, IEEE Journal of Biomedical and Health Informatics.

[18]  Ram Bilas Pachori,et al.  Automatic diagnosis of septal defects based on tunable-Q wavelet transform of cardiac sound signals , 2015, Expert Syst. Appl..

[19]  S. Mangione,et al.  Cardiac auscultatory skills of physicians-in-training: a comparison of three English-speaking countries. , 2001, The American journal of medicine.

[20]  Hayong Shin,et al.  Classification of heart sound recordings using convolution neural network , 2016, 2016 Computing in Cardiology Conference (CinC).

[21]  Harun Uguz,et al.  Adaptive neuro-fuzzy inference system for diagnosis of the heart valve diseases using wavelet transform with entropy , 2011, Neural Computing and Applications.

[22]  Qiang Huang,et al.  Attention and Localization Based on a Deep Convolutional Recurrent Model for Weakly Supervised Audio Tagging , 2017, INTERSPEECH.

[23]  Sridha Sridharan,et al.  Heart Sound Segmentation Using Bidirectional LSTMs With Attention , 2020, IEEE Journal of Biomedical and Health Informatics.

[24]  Goutam Saha,et al.  Detection of cardiac abnormality from PCG signal using LMS based least square SVM classifier , 2010, Expert Syst. Appl..

[25]  Takumi Kobayashi,et al.  Global Feature Guided Local Pooling , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[26]  Thomas G. Dietterich.  Approximate Statistical Tests for Comparing Supervised Classification Learning Algorithms , 1998, Neural Computation.

[27]  Björn W. Schuller,et al.  The INTERSPEECH 2009 emotion challenge , 2009, INTERSPEECH.

[28]  Björn W. Schuller,et al.  Learning Image-based Representations for Heart Sound Classification , 2018, DH.

[29]  Amina Adadi,et al.  Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI) , 2018, IEEE Access.

[30]  Geoffrey E. Hinton,et al.  Layer Normalization , 2016, ArXiv.

[31]  Björn Schuller,et al.  Wavelets Revisited for the Classification of Acoustic Scenes , 2017, DCASE.

[32]  Kavita Radhakrishnan,et al.  Recommendations for the Implementation of Telehealth in Cardiovascular and Stroke Care: A Policy Statement From the American Heart Association , 2017, Circulation.

[33]  Manop Phankokkruad,et al.  A Comparison of Efficiency Improvement for Long Short-Term Memory Model Using Convolutional Operations and Convolutional Neural Network , 2019, 2019 International Conference on Information and Communications Technology (ICOIACT).

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

[35]  Mark D. Plumbley,et al.  Attention-based Atrous Convolutional Neural Networks: Visualisation and Understanding Perspectives of Acoustic Scenes , 2019, ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[36]  Christian Biemann,et al.  What do we need to build explainable AI systems for the medical domain? , 2017, ArXiv.

[37]  Sepp Hochreiter,et al.  The Vanishing Gradient Problem During Learning Recurrent Neural Nets and Problem Solutions , 1998, Int. J. Uncertain. Fuzziness Knowl. Based Syst..

[38]  Kun Qian,et al.  Can Machine Learning Assist Locating the Excitation of Snore Sound? A Review , 2020, IEEE Journal of Biomedical and Health Informatics.

[39]  Björn W. Schuller,et al.  Deep Unsupervised Representation Learning for Abnormal Heart Sound Classification , 2018, 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[40]  Björn Schuller,et al.  Sequence to Sequence Autoencoders for Unsupervised Representation Learning from Audio , 2017, DCASE.

[41]  Björn W. Schuller,et al.  Active Learning by Sparse Instance Tracking and Classifier Confidence in Acoustic Emotion Recognition , 2012, INTERSPEECH.

[42]  Ping Wang,et al.  Phonocardiographic Signal Analysis Method Using a Modified Hidden Markov Model , 2007, Annals of Biomedical Engineering.

[43]  Takio Kurita,et al.  Improvement of learning for CNN with ReLU activation by sparse regularization , 2017, 2017 International Joint Conference on Neural Networks (IJCNN).

[44]  Kun Qian,et al.  Deep Scalogram Representations for Acoustic Scene Classification , 2018, IEEE/CAA Journal of Automatica Sinica.

[45]  Shi-Wen Deng,et al.  Towards heart sound classification without segmentation via autocorrelation feature and diffusion maps , 2016, Future Gener. Comput. Syst..