Abstract:Phonocardiogram (PCG) plays an important role in evaluating many cardiac abnormalities, such as the valvular heart disease, congestive heart failure and anatomical defects of the heart. However, effective cardiac auscultation requires trained physicians whose work is tough, laborious and subjective. The objective of this study is to develop an automatic classification method for anomaly (normal vs. abnormal) detection of PCG recordings without any segmentation of heart sound signals. Hybrid signal processing and artificial intelligence tools, including tunable Q-factor wavelet transform (TQWT), variational mode decomposition (VMD), phase space reconstruction (PSR) and neural networks, are utilized to extract representative features in order to model, identify and detect abnormal patterns in the dynamics of PCG system caused by heart disease. First, heart sound signal is decomposed into a set of frequency subbands with a number of decomposition levels by using the TQWT method. Second, VMD is employed to decompose the subband of the heart sound signal into different intrinsic modes, in which the first four intrinsic modes contain the majority of the heart sound signal's energy and are considered to be the predominant intrinsic modes. They are selected to construct the reference variable for analysis. Third, phase space of the reference variable is reconstructed, in which the properties associated with the nonlinear PCG system dynamics are preserved. Three-dimensional PSR together with Euclidean distance has been utilized to derive features, which demonstrate significant difference in PCG system dynamics between normal and abnormal heart sound signals. Finally, PhysioNet/CinC Challenge heart sound database is used for evaluation and the synthetic minority over-sampling technique method is applied to balance the datasets. By using the 10-fold cross-validation style, experimental results demonstrate that the proposed features with dynamical neural networks based classifier yield classification performance with sensitivity, specificity, overall score and accuracy values of 97.73\(\%\), 98.05\(\%\), 97.89\(\%\), and 97.89\(\%\), respectively. The results verify the effectiveness of the proposed method which can serve as a potential candidate for the automatic anomaly detection in the clinical application.

Imbalanced Heart Sound Signal Classification Based on Two-Stage Trained DsaNet

HADLN: Hybrid Attention-Based Deep Learning Network for Automated Arrhythmia Classification

AmtNet: Attentional multi-scale temporal network for phonocardiogram signal classification

Automated heart sound classification system from unsegmented phonocardiogram (PCG) using deep neural network

Pay More Attention with Fewer Parameters: A Novel 1-D Convolutional Neural Network for Heart Sounds Classification

Automatic Classification of Normal–Abnormal Heart Sounds Using Convolution Neural Network and Long-Short Term Memory

Heart Sound Signals Classification with Image Conversion Employed

Multi-label Classification of Abnormalities in 12-Lead ECG Using Deep Learning.

A novel heart sound segmentation algorithm via multi-feature input and neural network with attention mechanism

Automated Heartbeat Classification Exploiting Convolutional Neural Network with Channel-Wise Attention

Real-time PCG Anomaly Detection by Adaptive 1D Convolutional Neural Networks

A learnable front-end based efficient channel attention network for heart sound classification

A new approach for the detection of abnormal heart sound signals using TQWT, VMD and neural networks

HBNET: A blended ensemble model for the detection of cardiovascular anomalies using phonocardiogram

ECG signal classification based on deep CNN and BiLSTM

A Novel Imbalanced Dataset Mitigation Method and ECG Classification Model Based on Combined 1D_cbam-Autoencoder and Lightweight CNN Model.

Multi-class 12-lead ECG automatic diagnosis based on a novel subdomain adaptive deep network

DECNet: Left Atrial Pulmonary Vein Class Imbalance Classification Network

A Robust Deep Learning Framework Based on Spectrograms for Heart Sound Classification

Transfer Learning Models for Detecting Six Categories of Phonocardiogram Recordings

Phonocardiogram Classification by Learning From Positive and Unlabeled Examples