Automatic Heart Sound Classification Research Articles

Hybrid Sneaky algorithm-based deep neural networks for Heart sound classification using phonocardiogram

ABSTRACT In the diagnosis of cardiac disorders Heart sound has a major role, and early detection is crucial to safeguard the patients. Computerized strategies of heart sound classification advocate intensive and more exact results in a quick and better manner. Using a hybrid optimization-controlled deep learning strategy this paper proposed an automatic heart sound classification module. The parameter tuning of the Deep Neural Network (DNN) classifier in a satisfactory manner is the importance of this research which depends on the Hybrid Sneaky optimization algorithm. The developed sneaky optimization algorithm inherits the traits of questing and societal search agents. Moreover, input data from the Phonocardiogram (PCG) database undergoes the process of feature extraction which extract the important features, like statistical, Heart Rate Variability (HRV), and to enhance the performance of this model, the features of Mel frequency Cepstral coefficients (MFCC) are assisted. The developed Sneaky optimization-based DNN classifier’s performance is determined in respect of the metrics, namely precision, accuracy, specificity, and sensitivity, which are around 97%, 96.98%, 97%, and 96.9%, respectively.

Early-warning of Cardiac Condition through Detection of Murmur in Heart Sound - A Case Study.

Cardiovascular disease, particularly Rheumatic Heart Disease (RHD), is one of the leading causes of death in many developing countries. RHD is manageable and treatable with early detection. However, multiple countries across the globe suffer from a scarcity of experienced physicians who can perform screening at large scales. Advancements in machine learning and signal processing have paved way for Phonocardiogram (PCG)-based automatic heart sound classification. The direct implication of such methods is that it is possible to enable a person without specialized training to detect potential cardiac conditions with just a digital stethoscope. Hospitalization or life-threatening situations can be dramatically reduced via such early screenings. Towards this, we conducted a case study amongst a population from a particular geography using machine learning and deep learning methods for the detection of murmur in heart sounds. The methodology consists of first pre-processing and identifying normal vs. abnormal heart sound signals using 3 state-of-the-art methods. The second step further identifies the murmur to be systolic or diastolic by capturing the auscultation location. Abnormal findings are then sent for early attention of clinicians for proper diagnosis. The case study investigates the efficacy of the automated method employed for early screening of potential RHD and initial encouraging results of the study are presented.

Research process on deep learning methods for heart sounds classification

Cardiovascular diseases are still the primary threats to people’s health around the world. Automatic heart sound classification technology, as a fast and efficient means for diagnosis and treatment, is of great clinical significance. With the rapid development of artificial intelligence technology, deep learning algorithms are widely used in automatic heart sound classification. This paper reviewed the key technologies related to the automatic classification of heart sounds in recent years, including heart sound denoising, segmentation, feature extraction, and classification recognition. The classification and recognition technologies related to deep learning are presented in detail, with a focus on the application and development of convolutional neural network and recurrent neural network, as well as various combination models for heart sound classification in the past five years.

Corrigendum to: Novel complex-valued deep learning applied to automatic classification of heart sounds

Corrigendum to: Novel complex-valued deep learning applied to automatic classification of heart sounds

Novel complex-valued deep learning applied to automatic classification of heart sounds

Abstract Introduction Cardiovascular disease (CVD) is a problem facing governments around the world. Early detection saves the healthcare system thousands of dollars by allowing prevention and avoiding costly treatments when the disease has progressed to advanced stages. According to the World Health Organization (WHO), cardiovascular diseases represent the leading cause of mortality worldwide (WHO, 2019). That is why, we have decided to develop novel artificial intelligence models for the detection of heart disease, with accuracies that allow us to put the obtained algorithms into production. The revolution of artificial intelligence allows the application Deep Learning techniques of two-dimensional images in the domain of both real numbers and complex value numbers, as classifiers of heart sounds, for the detection of normality or abnormality in the functioning of the heart. Purpose In the present work, we propose the comparison of a novel 2D convolutional neural network (CNN) algorithm in the domain of complex value numbers with its counterpart in the domain of real numbers for the automatic classification of heart sounds into normal or abnormal. Material The database we decided to use for our research is Pascal, which has audio files of cardiac activity, distributed in 351 normal sounds and 129 pathological sounds. Methods The following steps were applied to get the objectives of our work: 1) automatic segmentation of a single heartbeat, 2) conversion of the segmented sound into its associated image scalogram using the Hilbert transform, 3) classification of the sounds into normal and abnormal using the proposed algorithms, and 4) measurement and comparison of the results obtained by performing a two-tailed t-student hypothesis test and cross-validation. Results We present a comparative table between the two proposed models, finding that Accuracy, F1 Score, Precision and Recall metrics obtained using complex-valued convolution networks present significant improvements compared with the real valued one. The following table show us the obtained numbers. For all cases, the t-student test shows us p-values less than 0.05%, giving statistical evidence that the means are significantly different between the two proposed models. Besides, in all cases, the performance of the Complex-valued model is better compared with the Real-valued one. Conclusion Complex-valued neural networks propose a significant advance around Deep learning, since they present a better performance than the traditional counterparts based on real numbers. This proposes an experimental basis for the construction of a new Deep learning paradigm, where information in another numerical domain, is better exploited with the help of mathematical transforms. The latter is a significant advance in health sciences, where the demand is higher, in terms of performance of the proposed models. Funding Acknowledgement Type of funding sources: Other. Main funding source(s): eVIDA research Group

Deep attention-based neural networks for explainable heart sound classification

Cardiovascular diseases are the leading cause of death and severely threaten human health in daily life. There have been dramatically increasing demands from both the clinical practice and the smart home application for monitoring the heart status of individuals suffering from chronic cardiovascular diseases. However, experienced physicians who can perform efficient auscultation are still lacking in terms of number. Automatic heart sound classification leveraging the power of advanced signal processing and deep learning technologies has shown encouraging results. Nevertheless, a lack of explanation for deep neural networks is a limitation for the applications of automatic heart sound classification. To this end, we propose explaining deep neural networks for heart sound classification with an attention mechanism. We evaluate the proposed approach on the heart sounds shenzhen corpus. Our approach achieves an unweighted average recall of 51.2% for classifying three categories of heart sounds, i.e., normal, mild, and moderate/severe. The experimental results also demonstrate that the global attention pooling layer improves the performance of the learnt representations by estimating the contribution of each unit in high-level features. We further analyse the deep neural networks by visualising the attention tensors.

Heart Sound Analysis for Abnormality Detection

According to the World Health Organization, 31% death rate in the World is because of cardiovascular diseases like heart arrhythmia and heart failure. Early diagnosis of heart problems may help in timely treatment of the patients and hence control death rate. Heart sounds are good signals of heart health if examined by an expert. Moreover, heart sounds can be analyzed with inexpensive and portable medical devices. Automatic heart sound classification can be very useful in diagnosing heart problems. Major focus of this research is to study the existing techniques for heart sound classification and develop a more sophisticated method. A signal processing technique is been proposed for heart sound classification. Five classifiers, Naive Bayes algorithm, Sequential (SMO), J48, Rep tree and Random Forest (RF) are used for this experiment. A detailed experimentation is performed to fine-tune the method and finally results are compared with the existing systems. The best proposed classifying technique results the overall accuracy of 91.33%.

Transferring Cross-Corpus Knowledge: An Investigation on Data Augmentation for Heart Sound Classification.

Human auscultation has been regarded as a cheap, convenient and efficient method for the diagnosis of cardiovascular diseases. Nevertheless, training professional auscultation skills needs tremendous efforts and is time-consuming. Computer audition (CA) that leverages the power of advanced machine learning and signal processing technologies has increasingly attracted contributions to the field of automatic heart sound classification. While previous studies have shown promising results in CA based heart sound classification with the 'shuffle split' method, machine learning for heart sound classification decreases in accuracy with a cross-corpus test dataset. We investigate this problem with a cross-corpus evaluation using the PhysioNet CinC Challenge 2016 Dataset and propose a new combination of data augmentation techniques that leads to a CNN robust for such cross-corpus evaluation. Compared with the baseline, which is given without augmentation, our data augmentation techniques combined improve by 20.0 % the sensitivity and by 7.9 % the specificity on average across 6 databases, which is a significant difference on 4 out of these (p < .05 by one-tailed z-test).

Heart sound classification based on sub-band envelope and convolution neural network

Automatic classification of heart sounds plays an important role in the early diagnosis of congenital heart disease. A kind of heart sound classification algorithms based on sub-band envelope feature and convolution neural network was proposed in this paper, which did not need to segment the heart sounds according to cardiac cycle accurately. Firstly, the heart sound signal was divided into some frames. Then, the frame level heart sound signal was filtered with Gammatone filter bank to obtain the sub-band signals. Next, the sub-band envelope was extracted by Hilbert transform. After that, the sub-band envelope was stacked into a feature map. Finally, type Ⅰ and type Ⅱ convolution neural network were selected as classifier. The result shown that the sub-band envelope feature was better in type Ⅰ than type Ⅱ. The algorithm is tested with 1 000 heart sound samples. The test results show that the overall performance of the algorithm proposed in this paper is significantly improved compared with other similar algorithms, which provides a new method for automatic classification of congenital heart disease, and speeds up the process of automatic classification of heart sounds applied to the actual screening.

An Efficacy of Spectral Features with Boosted Decision Tree Algorithm for Automatic Heart Sound Classification

This research work aims to classify the audio signals received from heart into normal/abnormal. The heart sound perceived has been referred as phonocardiogram (PCG) signals. An attempt has been made to identify a set of features that provide more accurate results for classifying PCG under designated categories using a variant of decision tree algorithm. After applying 6th order butter worth band-pass filter on PCG signals, the new features, viz. Tonnetz, Spectral contrast, and Chroma have been extracted. Further, XGBOOST, a variant of the decision tree has been used for classifying unsegmented PCG signals. The benchmark datasets, PhysioNet 2016, and PASCAL 2011 have been taken for validating the proposed methodology presented here. PhysioNet 2016 is comprised of sub-datasets, viz. A–F which contain a total of 3,240 PCG recordings, whereas the PASCAL 2011 contains 415 heart sound signals. The proposed approach considers a new feature set in conjunction with the existing ones; and it has resulted in mean accuracy, sensitivity, and specificity scores as 95.2, 94.22 and 96.18 respectively.

Lightweight End-to-End Neural Network Model for Automatic Heart Sound Classification

Heart sounds play an important role in the initial screening of heart diseases. However, the accurate diagnosis with heart sound signals requires doctors to have many years of clinical experience and relevant professional knowledge. In this study, we proposed an end-to-end lightweight neural network model that does not require heart sound segmentation and has very few parameters. We segmented the original heart sound signal and performed a short-time Fourier transform (STFT) to obtain the frequency domain features. These features were sent to the improved two-dimensional convolutional neural network (CNN) model for features learning and classification. Considering the imbalance of positive and negative samples, we introduced FocalLoss as the loss function, verified our network model with multiple random verifications, and, hence, obtained a better classification result. Our main purpose is to design a lightweight network structure that is easy for hardware implementation. Compared with the results of the latest literature, our model only uses 4.29 K parameters, which is 1/10 of the size of the state-of-the-art work.

Heart Sounds Classification Based on Feature Fusion Using Lightweight Neural Networks

Heart sounds are an important basis for evaluating heart disease. In recent years, auxiliary diagnosis technology of cardiovascular disease based on the detection of heart sound signals has become a research hotspot. A lightweight automatic heart sound classification method is presented in this study. It only needs to perform simple preprocessing on the heart sound data, and a neural network model is constructed to automatically extract the time–frequency features of heart sound data: where the convolution module extracts the frequency-domain characteristics, the loop module extracts the time-domain characteristics, and the features are then fused in series and parallel. Finally, the classification and recognition of heart sounds are realized based on the fusion features. At the same time, methods, such as group convolution, global average pooling, and gated loop mechanisms, are used to reduce the parameters and training time of the neural network. The experimental data from the PhysioNet database are used for testing. The recognition accuracy of the series feature fusion heart sound classification model is 95.00%, and the model size is 0.6 MB. In addition, the parallel feature fusion heart sound classification model is 95.50%, and the model size is 1.36 MB. Therefore, the lightweight heart sound classification model designed in this study contains fewer parameters and higher recognition accuracy. This makes it suitable for deployment in embedded devices. It also has important research significance for the development of portable heart sound detection equipment.

Automatic heart sound classification from segmented/unsegmented phonocardiogram signals using time and frequency features

Objective: Heart abnormality detection using heart sound signals (phonocardiogram (PCG)) has been an active research area for the last few decades. In this paper, automatic heart sound classification using segmented and unsegmented PCG signals is presented. Approach: In this paper: (i) we perform an in-depth analysis of various time and frequency domain features, followed by experimental determination of effective feature subsets for improved classification performance; (ii) both segmented and unsegmented PCG signals are studied and important results concerning the respective feature subsets and their classification performances are reported; and (iii) different classification algorithms, including the support vector machine, kth nearest neighbor, decision tree, ensemble classifier, artificial neural network and long short-term memory network (LSTMs), are employed to evaluate the performance of the proposed feature subsets and their comparison with other established features and methods is presented. Main results: It is observed that LSTM performs better on mel-frequency cepstral coefficient (MFCC) features extracted from unsegmented PCG data, with an area under curve (AUC) score of 91.39%, however, the MFCC features do not show a consistent performance with other classifiers (the second highest AUC score is 62.08% with the decision tree classifier). In contrast, in the case of time-frequency features from segmented data, the performance of all the classifiers is appreciable with AUC scores over 70%. In particular, the conventional machine learning techniques shows consistency in achieving over 80% in AUC scores. Significance: The results of this study highlight the importance of time and frequency domain features. Thus it is necessary to employ both the time and frequency features of segmented PCG signals to achieve improved classification.

Cardiac valves disorder classification based on active valves appearance periodic sequences tree of murmurs

Heart sound auscultation and interpretation is a primary diagnosis and cardiac physical examination method. Heart sound classification becomes a challenging issue in the present of complicated heart sound or multiple valvular heart diseases. Unlike the existence automatic heart sound classification techniques, a powerful heart sound classification method utilizing spatial-temporal information is proposed to determine the heart murmur type when multiple heart murmurs are occurred. First of all, normal and abnormal heart sounds are discriminated via power spectra density of heart sound feature extraction along with Support Vector Machine classifier. Spatial-temporal information of heart valves are extracted by microphone array recording and recursively sound source localization with high resolution. A tree of active heart valves appearance sequences of heart murmurs are obtained by mining locations of active heart valves to model heart murmurs. Due to the heart cyclostationary nature, all sequences of this murmur tree are periodic. Normal heart sound and abnormal heart sound containing various valve regurgitation and stenosis are collected by a rectangular array of 2 × 3 microphones. To evaluate the benefits of the proposed method, it is compared with a murmur classification method based on power spectra density feature extraction. The proposed method achieves 89.8 % accuracy which is superior to mentioned method.

A study of time-frequency features for CNN-based automatic heart sound classification for pathology detection

This study concerns the task of automatic structural heart abnormality risk detection from digital phonocardiogram (PCG) signals aiming at pediatric heart disease screening applications. Recently, various systems based on convolutional neural networks trained on time-frequency representations of segmental PCG frames have been presented that outperform systems using hand-crafted features. This study focuses on the segmentation and time-frequency representation components of the CNN-based designs. We consider the most commonly used features (MFCC and Mel-Spectrogram) used in state-of-the-art systems and a time-frequency representation influenced by domain-knowledge, namely sub-band envelopes as an alternative feature. Via tests carried on two high quality databases with a large set of possible settings, we show that sub-band envelopes are preferable to the most commonly used features and period synchronous windowing is preferable over asynchronous windowing.

Heart sound classification based on scaled spectrogram and tensor decomposition

First, the spectrograms of heart cycles are scaled for comparison.Second, tensor decomposition is utilized to the scaled spectrograms.Third, the intrinsic structure information of scaled spectrograms is extracted.Fourth, more useful physiological and pathological information is reserved.Fifth, the extracted features are more discriminative. Heart sound signal analysis is an effective and convenient method for the preliminary diagnosis of heart disease. However, automatic heart sound classification is still a challenging problem which mainly reflected in heart sound segmentation and feature extraction from the corresponding segmentation results. In order to extract more discriminative features for heart sound classification, a scaled spectrogram and tensor decomposition based method was proposed in this study. In the proposed method, the spectrograms of the detected heart cycles are first scaled to a fixed size. Then a dimension reduction process of the scaled spectrograms is performed to extract the most discriminative features. During the dimension reduction process, the intrinsic structure of the scaled spectrograms, which contains important physiological and pathological information of the heart sound signals, is extracted using tensor decomposition method. As a result, the extracted features are more discriminative. Finally, the classification task is completed by support vector machine (SVM). Moreover, the proposed method is evaluated on three public datasets offered by the PASCAL classifying heart sounds challenge and 2016 PhysioNet challenge. The results show that the proposed method is competitive.

Classification of heart sounds using empirical mode decomposition based features

A novel method is presented to extract robust features for automatic classification of heart sounds based on Empirical Mode Decomposition (EMD). The work decomposes segmented heart sound cycles with EMD to generate certain intrinsic mode functions (IMFs). It is seen that the first IMF contains mostly high frequency noise, the second and third IMFs carry higher frequency components of our signal of interest and residue contains its low frequency components. A twenty five dimensional feature vector is generated from average energy of the segmented IMFs and residue which serve as input to classifier models. Two different classifiers, Artificial Neural Network (ANN) and Grow and Learn (GAL) network, are used to show the performance of the proposed feature extraction technique. Experiments are conducted on 104 different recordings of heart sound comprising of normal and 12 different pathological cases against three different additive background noises – white Gaussian, hospital and body noise. It is found that the EMD based feature extraction always performs better than benchmark wavelet based feature extraction technique.

Wavelet-based Segmentation and Feature Extraction of Heart Sounds for Intelligent PDA-based Phonocardiography

Many pathological conditions of the cardiovascular system cause murmurs and aberrations in heart sounds. Phonocardiography provides the clinician with a complementary tool to record the heart sounds heard during auscultation. The advancement of intracardiac phonocardiography combined with modern digital signal processing techniques has strongly renewed researchers' interest in studying heart sounds and murmurs. The aim of this work is to investigate the applicability of different spectral analysis methods to heart sound signals and explore their suitability for PDA-based implementation. Fourier transform (FT), short-time Fourier transform (STFT) and wavelet transform (WT) are used to perform spectral analysis on heart sounds. A segmentation algorithm based on Shannon energy is used to differentiate between first and second heart sounds. Then wavelet transform is deployed again to extract 64 features of heart sounds. The FT provides valuable frequency information but the timing information is lost during the transformation process. The STFT or spectrogram provides valuable time-frequency information but there is a trade-off between time and frequency resolution. Wavelet analysis, however, does not suffer from limitations of the STFT and provides adequate time and frequency resolution to accurately characterize the normal and pathological heart sounds. The results show that the wavelet-based segmentation algorithm is quite effective in localizing the important components of both normal and abnormal heart sounds. They also demonstrate that wavelet-based feature extraction provides suitable feature vectors which are clearly differentiable and useful for automatic classification of heart sounds.