Heart Sound Signals Research Articles

Sports Training Health Analysis Algorithm Based on Heart Rhythm Feature Extraction and Convolutional Neural Network.

Intelligent sports equipment and software have emerged in the field of sports as a result of the advancement of information technology, allowing professional athletes to collect and visually display the movement and physical signs of the human body to aid in the planning of sports strategies. Intuitive data, on the other hand, cannot assist ordinary people who lack professional knowledge in exercising correctly. As a result, in the field of intelligent sports and health, effective use of collected exercise and physical sign data to analyze the user's personal physical condition and generate reasonable exercise suggestions has emerged as a research direction. In humans, the heart sound signal is a biological signal. It can help people detect and monitor heart health problems by analyzing the characteristics of heart sound signals. The goal of this paper is to use heart sound to identify and analyze athletes' training health. It provides a revolutionary health analysis algorithm based on heart rhythm feature extraction and convolutional neural networks, which is based on exercise training. It greatly improves the accuracy of the recognition and prediction of the athlete's training health status.

Prediction of exercise sudden death in rabbit exhaustive swimming using deep neural network

Background and objectiveModerate exercise contributes to good health. However, excessive exercise may lead to cardiac fatigue, myocardial damage and even exercise sudden death. Monitoring the heart health has important implication to prevent exercise sudden death. Diagnosis methods such as electrocardiogram, echocardiogram, blood pressure and histological analysis have shown that arrhythmia and left ventricular fibrosis are early warning symptoms of exercise sudden death. Heart sounds (HS) can reflect the changes of cardiac valve, cardiac blood flow and myocardial function. Deep learning has drawn wide attention because of its ability to recognize disease. Therefore, a deep learning method combined with HS was proposed to predict exercise sudden death in New Zealand rabbits. The objective is to develop a method to predict exercise sudden death in New Zealand rabbits.MethodsThis paper proposed a method to predict exercise sudden death in New Zealand rabbits based on convolutional neural network (CNN) and gated recurrent unit (GRU). The weight-bearing exhaustive swimming experiment was conducted to obtain the HS of exercise sudden death and surviving New Zealand rabbits (n = 11/10) at four different time points. Then, the improved Viola integral method and double threshold method were employed to segment HS signals. The segmented HS frames at different time points were taken as the input of a combined CNN and GRU called CNN–GRU network to complete the prediction of exercise sudden death.ResultsIn order to evaluate the performance of proposed network, CNN and GRU were used for comparison. When the fourth time point segmented HS frames were taken as input, the result shows that the proposed network has better performance with an accuracy of 89.57%, a sensitivity of 89.38% and a specificity of 92.20%. In addition, the segmented HS frames at different time points were input into CNN–GRU network, and the result shows that with the progress of the experiment, the prediction accuracy of exercise sudden death in New Zealand rabbits increased from 50.98 to 89.57%.ConclusionThe proposed network shows good performance in classifying HS, which proves the feasibility of deep learning in exploring exercise sudden death. Further, it may have important implications in helping humans explore exercise sudden death.

Classification of heart sounds based on quality assessment and wavelet scattering transform

Automatic classification of heart sound plays an important role in the diagnosis of cardiovascular diseases. In this study, a heart sound sample classification method based on quality assessment and wavelet scattering transform was proposed. First, the ratio of zero crossings (RZC) and the root mean square of successive differences (RMSSD) were used for assessing the quality of heart sound signal. The first signal segment conforming to the threshold standard was selected as the current sample for the continuous heart sound signal. Using the wavelet scattering transform, the wavelet scattering coefficients were expanded according to the wavelet scale dimension, to obtain the features. Support vector machine (SVM) was used for classification, and the classification results for the samples were obtained using the wavelet scale dimension voting approach. The effects of RZC and RMSSD on the results are discussed in detail. On the database of PhysioNet Computing in Cardiology Challenge 2016 (CinC 2016), the proposed method yields 92.23% accuracy (Acc), 96.62% sensitivity (Se), 90.65% specificity (Sp), and 93.64% measure of accuracy (Macc). The results show that the proposed method can effectively classify normal and abnormal heart sound samples with high accuracy.

Machine Learning based Heart Disease Diagnosis using Non-Invasive Methods: A Review

Heart disease is the most dangerous disease among all the non-communicable diseases. Annually 17900 thousand of peoples die due to heart problems. Cardiovascular disease (CVD) is the general term used for most of the heart diseases. There are two types of methods for diagnosing a CVD: (i) Invasive Methods (ii) Non-Invasive Methods. Coronary angiography is an invasive method for diagnosing a CVD which is a costly, painful and complicated process. A variety of Non-Invasive (NI) methods are available for diagnosing a CVD. NI methods generate a lot of data which is mainly of 3 kinds :(i) data based on clinical parameters, lab tests and symptoms (ii)data based on raw heart signals (ECG and PCG) (iii)data based on heart images. Majorly, three different machine learning (ML) frameworks may be developed based on the 3 types of data. First framework is simple and main concern is feature selection and classification. Second and third framework is complicated and requires a lot of techniques (preprocessing, segmentation and feature extraction) prior to classification of heart signals and images respectively. In this paper a comprehensive review is presented that summarizes some recent and prevalent machine learning methodologies in all the frameworks. Most of the papers reviewed in this study are from IEEE Explorer, Science Direct, PubMed, Springer, Hindawi, ACM digital library and MDPI libraries. It is found that Support Vector Machines (SVM) and Artificial Neural Networks (ANN) are superseding in most of the studies in all the frameworks. Deep neural network is comparatively newer machine learning methodology which is giving prominent results in classifying heart sound signals and cardiovascular images. The present study will help to automate diagnosis process of heart disease by providing guidelines and avenues to new researchers in domain of machine learning.

Design and verification of magnetic-induction electronic stethoscope based on MEMS technology

A magnetic-induction electronic stethoscope based on MEMS technology is proposed in this paper. The stethoscope, based on MEMS technology, combines the bionic principle, piezoresistive principle and magnetic induction principle. Through the design and optimization of the microstructure of the MEMS sensor, the sensitivity of the stethoscope to low-frequency heart sound signal detection is significantly improved (−189.0dB@500Hz), the signal-to-noise ratio reaches 27.38dB, which is 4.08dB higher than similar 3M Littmann 3200 electronic stethoscope products, and it meets the frequency band requirements of heart sound collection (20-600 Hz). The feasibility of the stethoscope for auscultation of heart sound is verified in this paper through theoretical analysis, finite element simulation and experimental testing. The results show that the electronic stethoscope features high sensitivity, strong anti-interference ability, better low-frequency characteristics, smaller size, and greater technical advantages than similar electronic stethoscope products.

Heart sound classification based on log Mel-frequency spectral coefficients features and convolutional neural networks

In view of the important role of heart sound signals in diagnosing and preventing congenital heart disease, a novel method about feature extraction and classification of heart sound signals was put forward in this study. Firstly, the heart sound signals were de-noised by using the wavelet algorithm. Subsequently, the improved duration-dependent hidden Markov model (DHMM) was used to segment the heart sound signal according to the heart cycle. Then, the dynamic frame length method was used to extract log Mel-frequency spectral coefficients (MFSC) features from the heart sound signal based on the heart cycle. Afterward, the convolution neural network (CNN) was used to classify the MFSC features. Finally, the majority voting algorithm was used to get the optimal classification results. In this paper, two-classification and multi-classification models were built. An accuracy of 93.89% for two-classification and an accuracy of 86.25% for multi-classification were achieved using the novel method.

Detection of Coronary Artery Disease Using Multi-Domain Feature Fusion of Multi-Channel Heart Sound Signals.

Heart sound signals reflect valuable information about heart condition. Previous studies have suggested that the information contained in single-channel heart sound signals can be used to detect coronary artery disease (CAD). But accuracy based on single-channel heart sound signal is not satisfactory. This paper proposed a method based on multi-domain feature fusion of multi-channel heart sound signals, in which entropy features and cross entropy features are also included. A total of 36 subjects enrolled in the data collection, including 21 CAD patients and 15 non-CAD subjects. For each subject, five-channel heart sound signals were recorded synchronously for 5 min. After data segmentation and quality evaluation, 553 samples were left in the CAD group and 438 samples in the non-CAD group. The time-domain, frequency-domain, entropy, and cross entropy features were extracted. After feature selection, the optimal feature set was fed into the support vector machine for classification. The results showed that from single-channel to multi-channel, the classification accuracy has increased from 78.75% to 86.70%. After adding entropy features and cross entropy features, the classification accuracy continued to increase to 90.92%. The study indicated that the method based on multi-domain feature fusion of multi-channel heart sound signals could provide more information for CAD detection, and entropy features and cross entropy features played an important role in it.

MACHINE-ASSISTED DIAGNOSIS OF CHD BASED ON VARIANT LOGIC THEORY AND MACHINE LEARNING MODELS

Cardiac auscultation is a basic means of initial diagnosis of congenital heart disease (CHD). It is significant to analyze Phonocardiogram (PCG) by using signal processing and machine learning techniques for the purpose of machine-assisted diagnosis of CHD. A novel method of machine-assisted diagnosis of CHD was proposed in this paper. First, the duration of the hidden Markov model (DHMM) was applied to locate and segment PCG into each cardiac cycle. Then, the envelope of the PCG was extracted by using Viola integral. After that, the variant logic theory was applied to extract the features and convert the envelope data of the heart sound signal into visual analysis measurement data. Finally, the classifier of the support vector machine (SVM) was used to classify the normal and abnormal heart sounds. There were 1000 cases used in this study. It was divided into a training set of 600 cases, a test set of 200 cases, and validation set of 200 cases. An accuracy of 0.965, a specificity of 0.898, and a sensitivity of 0.937 were achieved using the novel signal processing techniques. The results showed that DHMM and variant logic theory models were suitable for heart sound classification.

A FUSION FRAMEWORK BASED ON CEPSTRAL DOMAIN FEATURES FROM PHONOCARDIOGRAM TO PREDICT HEART HEALTH STATUS

A great increase in the number of cardiovascular cases has been a cause of serious concern for the medical experts all over the world today. In order to achieve valuable risk stratification for patients, early prediction of heart health can benefit specialists to make effective decisions. Heart sound signals help to know about the condition of heart of a patient. Motivated by the success of cepstral features in speech signal classification, authors have used here three different cepstral features, viz. Mel-frequency cepstral coefficients (MFCCs), gammatone frequency cepstral coefficients (GFCCs), and Mel-spectrogram for classifying phonocardiogram into normal and abnormal. Existing research has explored only MFCCs and Mel-feature set extensively for classifying the phonocardiogram. However, in this work, the authors have used a fusion of GFCCs with MFCCs and Mel-spectrogram, and achieved a better accuracy score of 0.96 with sensitivity and specificity scores as 0.91 and 0.98, respectively. The proposed model has been validated on the publicly available benchmark dataset PhysioNet 2016.

Heart sound recognition technology based on convolutional neural network

ABSTRACT The mortality rate of heart disease continues to rise each year: developing mechanisms to reduce mortality from heart disease is a top concern in today’s society. Heart sound auscultation is a crucial skill used to detect and diagnose heart disease. In this study, we propose a heart sound signal classification algorithm based on a convolutional neural network. The algorithm is based on heart sound data collected in the clinic and from medical books. The heart sound signals were first preprocessed into a grayscale image of 5 seconds. The training samples were then used to train and optimize the convolutional neural network; obtaining a training result with an accuracy of 95.17% and a loss value of 0.23. Finally, the convolutional neural network was used to test the test set samples. The results showed an accuracy of 94.80%, sensitivity of 94.29%, specificity of 95.54%, precision of 93.44%, F1_score of 93.84%, and an AUC of 0.943. Compared with other algorithms, the accuracy and sensitivity of the algorithms were improved. This shows that the method used in this study can effectively classify heart sound signals and could prove useful in assisting heart sound auscultation.

Detection and localization of S1 and S2 heart sounds by 3rd order normalized average Shannon energy envelope algorithm.

Adults born after 1970s are more prone to cardiovascular diseases. Death rate percentage is quite high due to heart related diseases. Therefore, there is necessity to enquire the problem or detection of heart diseases earlier for their proper treatment. As, Valvular heart disease, that is, stenosis and regurgitation of heart valve, are also a major cause of heart failure; which can be diagnosed at early-stage by detection and analysis of heart sound signal, that is, HS signal. In this proposed work, an attempt has been made to detect and localize the major heart sounds, that is, S1 and S2. The work in this article consists of three parts. Firstly, self-acquisition of Phonocardiogram (PCG) and Electrocardiogram (ECG) signal through a self-assembled, data-acquisition set-up. The Phonocardiogram (PCG) signal is acquired from all the four auscultation areas, that is, Aortic, Pulmonic, Tricuspid and Mitral on human chest, using electronic stethoscope. Secondly, the major heart sounds, that is, S1 and S2are detected using 3rd Order Normalized Average Shannon energy Envelope (3rd Order NASE) Algorithm. Further, an auto-thresholding has been used to localize time gates of S1 and S2 and that of R-peaks of simultaneously recorded ECG signal. In third part; the successful detection rate of S1 and S2, from self-acquired PCG signals is computed and compared. A total of 280 samples from same subjects as well as from different subjects (of age group 15-30 years) have been taken in which 70 samples are taken from each auscultation area of human chest. Moreover, simultaneous recording of ECG has also been performed. It was analyzed and observed that detection and localization of S1 and S2 found 74% successful for the self-acquired heart sound signal, if the heart sound data is recorded from pulmonic position of Human chest. The success rate could be much higher, if standard data base of heart sound signal would be used for the same analysis method. The, remaining three auscultations areas, that is, Aortic, Tricuspid, and Mitral have smaller success rate of detection of S1 and S2 from self-acquired PCG signals. So, this work justifies that the Pulmonic position of heart is most suitable auscultation area for acquiring PCG signal for detection and localization of S1 and S2 much accurately and for analysis purpose.

Application and Clinical Analysis of Remote Fetal Heart Rate Monitoring Platform in Continuous Fetal Heart Rate Monitoring Images.

Fetal heart sound is an important part of fetal monitoring and has attracted extensive research and attention from scholars at home and abroad in recent years. The fetal heart rate, extracted from the fetal heart sound signal, is one of the important indicators that reflect the health of the fetus in the womb. In this study, a maternal-fetal Holter monitor based on f ECG technology was used to collect maternal heart rate, fetal heart rate, and uterine contractions signals, isolate the fetal heart rate, and design an algorithm to extract the fetal heart rate baseline, acceleration, variation, wake-up cycle, and nonlinear parameters. Using statistical methods to analyze the average value and range of various characteristic parameters of fetal heart rate under continuous long-term monitoring, the results show that the baseline has a downward trend from 10 o'clock in the night to 4 o'clock in the morning and is the lowest around 2 o'clock in the morning. The area and acceleration time were significantly higher than those in the suspicious group. However, there was no significant difference in the number of acceleration values between the two groups; the proportion of small mutations in the normal group was lower than that of the suspicious group and the proportion of medium mutations was higher than that of the suspicious group. There is no statistically significant difference in maternal age, gestational age at childbirth, pregnancy comorbidities, and complication rates in the five-level interpretation system of ACOG (2009), RCOG (2007), SOGC (2007), and the United States (2007). The difference of pregnancy and parity in various images was statistically significant, P < 0.05. The second type of fetal heart rate monitoring images appeared in the highest among the diagnostic standards, and the difference in the second type of fetal heart rate monitoring images between the various diagnostic standards was statistically significant, P ≤ 0.001.

A heart sound classification method based on joint decision of extreme gradient boosting and deep neural network

Heart sound is one of the common medical signals for diagnosing cardiovascular diseases. This paper studies the binary classification between normal or abnormal heart sounds, and proposes a heart sound classification algorithm based on the joint decision of extreme gradient boosting (XGBoost) and deep neural network, achieving a further improvement in feature extraction and model accuracy. First, the preprocessed heart sound recordings are segmented into four status, and five categories of features are extracted from the signals based on segmentation. The first four categories of features are sieved through recursive feature elimination, which is used as the input of the XGBoost classifier. The last category is the Mel-frequency cepstral coefficient (MFCC), which is used as the input of long short-term memory network (LSTM). Considering the imbalance of the data set, these two classifiers are both improved with weights. Finally, the heterogeneous integrated decision method is adopted to obtain the prediction. The algorithm was applied to the open heart sound database of the PhysioNet Computing in Cardiology(CINC) Challenge in 2016 on the PhysioNet website, to test the sensitivity, specificity, modified accuracy and F score. The results were 93%, 89.4%, 91.2% and 91.3% respectively. Compared with the results of machine learning, convolutional neural networks (CNN) and other methods used by other researchers, the accuracy and sensibility have been obviously improved, which proves that the method in this paper could effectively improve the accuracy of heart sound signal classification, and has great potential in the clinical auxiliary diagnosis application of some cardiovascular diseases.

Automated Signal Quality Assessment for Heart Sound Signal by Novel Features and Evaluation in Open Public Datasets.

Automated heart sound signal quality assessment is a necessary step for reliable analysis of heart sound signal. An unavoidable processing step for this objective is the heart sound segmentation, which is still a challenging task from a technical viewpoint. In this study, ten features are defined to evaluate the quality of heart sound signal without segmentation. The ten features come from kurtosis, energy ratio, frequency-smoothed envelope, and degree of sound periodicity, where five of them are novel in signal quality assessment. We have collected a total of 7893 recordings from open public heart sound databases and performed manual annotation for each recording as gold standard quality label. The signal quality is classified based on two schemes: binary classification (“unacceptable” and “acceptable”) and triple classification (“unacceptable”, “good,” and “excellent”). Sequential forward feature selection shows that the feature “the degree of periodicity” gives an accuracy rate of 73.1% in binary SVM classification. The top five features dominate the classification performance and give an accuracy rate of 92%. The binary classifier has excellent generalization ability since the accuracy rate reaches to (90.4 ± 0.5) % even if 10% of the data is used to train the classifier. The rate increases to (94.3 ± 0.7) % in 10-fold validation. The triple classification has an accuracy rate of (85.7 ± 0.6) % in 10-fold validation. The results verify the effectiveness of the signal quality assessment, which could serve as a potential candidate as a preprocessing in future automatic heart sound analysis in clinical application.

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.

Rancang Bangun Phonocardiograf Berbasis Labview

The aim of this research is to develop a phonocardiograph system to display the signal and information form of the human heart sound examination. The analytical method applied is to use the Fast Fourier Transform (FFT) to determine the frequency of responses of human heart sound signals by using Labview software. The results obtained indicate the system able to display the signal shape of the heart sound, based on ten respondents whose data are taken shows the range of heart sound frequencies is in the range of 145 to 225 Hz with a duration of 81 to 120 bpm. Keywords: Phonocardiograph, FFT, Labview

Research on Segmentation and Classification of Heart Sound Signals Based on Deep Learning

The heart sound signal is one of the signals that reflect the health of the heart. Research on the heart sound signal contributes to the early diagnosis and prevention of cardiovascular diseases. As a commonly used deep learning network, convolutional neural network (CNN) has been widely used in images. In this paper, the method of analyzing heart sound through using CNN has been studied. Firstly, the original data set was preprocessed, and then the heart sounds were segmented on U-net, based on the deep CNN. Finally, the classification of heart sounds was completed through CNN. The data from 2016 PhysioNet/CinC Challenge was utilized for algorithm validation, and the following results were obtained. When the heart sound segmented, the overall accuracy rate was 0.991, the accuracy of the first heart sound was 0.991, the accuracy of the systolic period was 0.996, the accuracy of the second heart sound was 0.996, and the accuracy of the diastolic period was 0.997, and the average accuracy rate was 0.995; While in classification, the accuracy was 0.964, the sensitivity was 0.781, and the specificity was 0.873. These results show that deep learning based on CNN shows good performance in the segmentation and classification of the heart sound signal.

Automated Heart Sound Signal Segmentation and Identification using Abrupt Changes and Peak Finding Detection

Abstract Conventionally, majority physicians used manual-based approach to determine heart sound parameters by manually printing heart sound signal waveform on the paper and trace heart sound parameters duration. Towards modern medical diagnostic technologies, this paper presents a modified framework of automated heart sound signal segmentation and parameters identification for deployment in computer aided auscultation. Envelope-based approach with reference information by abrupt changes and peak detection algorithm are proposed for PASCAL Classifying Heart Sounds Challenge database. Noted that the performance of segmentation process was measured by calculating the accuracy and F1-score while the heart sound parameters were identified by computing its mean and standard deviation. The findings yield F1-score and accuracy of proposed approach for segmentation and identification of normal heart sound signal at 95.29% and 91.0%, respectively. Heart sound parameters such as first sound, S1 duration, second sound, S2 duration, systole duration, diastole duration, heart cycle duration and ratio of systole and diastole were also determined. This proposed approach is suitable to be applied on two prominent heart sound signal peaks of S1 and S2, which continuously varies among samples due to the different auscultatory sites.

A Bionic MEMS Electronic Stethoscope With Double-Sided Diaphragm Packaging

In this paper, an electronic stethoscope is designed based on a bionic Micro-Electro-Mechanical System (MEMS) sound-sensitive sensor. Inspired by the strong sound reflection effect of the water-air interface, a double-sided diaphragm MEMS electronic stethoscope (DMES) encapsulated by a novel double-sided diaphragm packaging is proposed. The double-sided diaphragm packaging’s superiority is verified by comparing a single diaphragm MEMS electronic stethoscope (SMES) with DMES. The frequency of the clinical heart sound signal is mostly in the range of 20–600 Hz. Finite simulation results show that for the same incident sound source, the sound pressure level inside double-sided diaphragm packaging is 4dB higher than that of single diaphragm packaging in this frequency band. Furthermore, the actual auscultation test results show that the signal-to-noise ratio (SNR) of DMES reaches 41.3dB, which is 2.2dB higher than SMES. Besides, comparing the heart sound signals collected by DMES and a commercial electronic stethoscope (Model 3200, 3M Littmann, USA), the high consistency of the two signals’ characteristic parameters in the time domain and frequency domain proves the feasibility of DMES. Finally, DMES has the advantages of low cost, and its SNR is 10.2db higher than that of the 3M electronic stethoscope. All these show that DMES has a broad prospect in the popularization of basic medical treatment.