Abnormal Heart Sounds Research Articles

A review on deep learning methods for heart sound signal analysis

IntroductionApplication of Deep Learning (DL) methods is being increasingly appreciated by researchers from the biomedical engineering domain in which heart sound analysis is an important topic of study. Diversity in methodology, results, and complexity causes uncertainties in obtaining a realistic picture of the methodological performance from the reported methods.MethodsThis survey paper provides the results of a broad retrospective study on the recent advances in heart sound analysis using DL methods. Representation of the results is performed according to both methodological and applicative taxonomies. The study method covers a wide span of related keywords using well-known search engines. Implementation of the observed methods along with the related results is pervasively represented and compared.Results and discussionIt is observed that convolutional neural networks and recurrent neural networks are the most commonly used ones for discriminating abnormal heart sounds and localization of heart sounds with 67.97% and 33.33% of the related papers, respectively. The convolutional neural network and the autoencoder network show a perfect accuracy of 100% in the case studies on the classification of abnormal from normal heart sounds. Nevertheless, this superiority against other methods with lower accuracy is not conclusive due to the inconsistency in evaluation.

Heart sound classification algorithm based on bispectral feature extraction and convolutional neural networks

Cardiovascular disease (CVD) is one of the leading causes of death worldwide. Heart sound classification plays a key role in the early detection of CVD. The difference between normal and abnormal heart sounds is not obvious. In this paper, in order to improve the accuracy of the heart sound classification model, we propose a heart sound feature extraction method based on bispectral analysis and combine it with convolutional neural network (CNN) to classify heart sounds. The model can effectively suppress Gaussian noise by using bispectral analysis and can effectively extract the features of heart sound signals without relying on the accurate segmentation of heart sound signals. At the same time, the model combines with the strong classification performance of convolutional neural network and finally achieves the accurate classification of heart sound. According to the experimental results, the proposed algorithm achieves 0.910, 0.884 and 0.940 in terms of accuracy, sensitivity and specificity under the same data and experimental conditions, respectively. Compared with other heart sound classification algorithms, the proposed algorithm shows a significant improvement and strong robustness and generalization ability, so it is expected to be applied to the auxiliary detection of congenital heart disease.

Fed-MStacking: Heterogeneous Federated Learning With Stacking Misaligned Labels for Abnormal Heart Sound Detection.

Ubiquitous sensing has been widely applied in smart healthcare, providing an opportunity for intelligent heart sound auscultation. However, smart devices contain sensitive information, raising user privacy concerns. To this end, federated learning (FL) has been adopted as an effective solution, enabling decentralised learning without data sharing, thus preserving data privacy in the Internet of Health Things (IoHT). Nevertheless, traditional FL requires the same architectural models to be trained across local clients and global servers, leading to a lack of model heterogeneity and client personalisation. For medical institutions with private data clients, this study proposes Fed-MStacking, a heterogeneous FL framework that incorporates a stacking ensemble learning strategy to support clients in building their own models. The secondary objective of this study is to address scenarios involving local clients with data characterised by inconsistent labelling. Specifically, the local client contains only one case type, and the data cannot be shared within or outside the institution. To train a global multi-class classifier, we aggregate missing class information from all clients at each institution and build meta-data, which then participates in FL training via a meta-learner. We apply the proposed framework to a multi-institutional heart sound database. The experiments utilise random forests (RFs), feedforward neural networks (FNNs), and convolutional neural networks (CNNs) as base classifiers. The results show that the heterogeneous stacking of local models performs better compared to homogeneous stacking.

Abnormal Heart Sound Detection using Time-Frequency Analysis and Machine Learning Techniques

Phonocardiogram (PCG) signals contains valuable information pertaining to heart valve functionality, rendering them potentially useful for early detection of cardiovascular diseases. Automated classification of heart sounds harbors great promise for identifying cardiac pathologies. This paper introduces a novel automated approach to classify normal and abnormal heart sounds. Our methodology involves partitioning heart sounds into four segments: S1, S2, systolic, and diastolic, followed by extraction of time–frequency and time-statistical features. Prior to data classification, we employ two techniques - particle Swarm optimization (PSO) and Sequential Forward Feature Selection (SFFS) - for efficient feature selection. We assess the performance of the proposed method on the Physionet Challenge 2016 database, utilizing the 10-fold cross-validation method. To address the issue of dataset imbalance, we apply the synthetic minority over-sampling technique (SMOTE) to create balanced datasets. Our approach surpasses existing methods in the literature, as evidenced by its superior accuracy, sensitivity, and specificity metrics. Specifically, our method achieves an accuracy of 98.03%, a sensitivity of 97.64%, and a specificity of 98.43% in distinguishing normal from abnormal heart sounds on the Physionet database. These findings outperform the results obtained by previously established methods evaluated on the Physionet 2016 challenge database.

Federated Abnormal Heart Sound Detection with Weak to No Labels.

Cardiovascular diseases are a prominent cause of mortality, emphasizing the need for early prevention and diagnosis. Utilizing artificial intelligence (AI) models, heart sound analysis emerges as a noninvasive and universally applicable approach for assessing cardiovascular health conditions. However, real-world medical data are dispersed across medical institutions, forming "data islands" due to data sharing limitations for security reasons. To this end, federated learning (FL) has been extensively employed in the medical field, which can effectively model across multiple institutions. Additionally, conventional supervised classification methods require fully labeled data classes, e.g., binary classification requires labeling of positive and negative samples. Nevertheless, the process of labeling healthcare data is time-consuming and labor-intensive, leading to the possibility of mislabeling negative samples. In this study, we validate an FL framework with a naive positive-unlabeled (PU) learning strategy. Semisupervised FL model can directly learn from a limited set of positive samples and an extensive pool of unlabeled samples. Our emphasis is on vertical-FL to enhance collaboration across institutions with different medical record feature spaces. Additionally, our contribution extends to feature importance analysis, where we explore 6 methods and provide practical recommendations for detecting abnormal heart sounds. The study demonstrated an impressive accuracy of 84%, comparable to outcomes in supervised learning, thereby advancing the application of FL in abnormal heart sound detection.

Detection method of early heart valve diseases based on heart sound features

Heart valve disease (HVD) is one of the common cardiovascular diseases. Heart sound is an important physiological signal for diagnosing HVDs. This paper proposed a model based on combination of basic component features and envelope autocorrelation features to detect early HVDs. Initially, heart sound signals lasting 5 minutes were denoised by empirical mode decomposition (EMD) algorithm and segmented. Then the basic component features and envelope autocorrelation features of heart sound segments were extracted to construct heart sound feature set. Then the max-relevance and min-redundancy (MRMR) algorithm was utilized to select the optimal mixed feature subset. Finally, decision tree, support vector machine (SVM) and k-nearest neighbor (KNN) classifiers were trained to detect the early HVDs from the normal heart sounds and obtained the best accuracy of 99.9% in clinical database. Normal valve, abnormal semilunar valve and abnormal atrioventricular valve heart sounds were classified and the best accuracy was 99.8%. Moreover, normal valve, single-valve abnormal and multi-valve abnormal heart sounds were classified and the best accuracy was 98.2%. In public database, this method also obtained the good overall accuracy. The result demonstrated this proposed method had important value for the clinical diagnosis of early HVDs.

Abnormal Heart Sound Classification and Model Interpretability: A Transfer Learning Approach with Deep Learning

Physician detection of heart sound abnormality is complicated by the inherent difficulty of detecting critical abnormalities in the presence of noise. Computer-aided heart auscultation provides a promising alternative for more accurate detection, with recent deep learning approaches exceeding expert accuracy. Although combining phonocardiogram (PCG) data with electrocardiogram (ECG) data provides more information to an abnormal heart sound classifier, the scarce presence of labelled datasets with this combination impedes training. This paper explores fine-tuning deep convolutional neural networks such as ResNet, VGG, and inceptionv3, on images of spectrograms, mel-spectrograms, and scalograms. By fine-tuning deep pre-trained models on image representations of ECG and PCG, we achieve 91.25% accuracy on the training-a dataset of the PhysioNet Computing in Cardiology Challenge 2016, compared to a previous result of 81.48%. Interpretation of the model’s learned features is also provided, with the results indicative of clinical significance.

Classification of Normal and Abnormal Heart Sounds Using Hjorth Features and LSTM – RNN Algorithm

The stethoscope is a commonly used medical device for diagnosing a patient’s condition by listening to their heartbeat. This diagnostic technique is known as auscultation, where each heart sound heard through the stethoscope has a distinct pattern that depends on the person’s heart condition. Researchers have developed computational methods that automatically analyze heart sounds to overcome the subjective nature of auscultation. In this study, we utilized the Hjorth method for feature extraction and the Long Short-Term RNN algorithm for classifying the normal and abnormal heart sounds. With 100 epochs, two layers of LSTM-RNN, and one layer of Dense, the classification accuracy for distinguishing normal and abnormal sounds reached 71.95%. This research aims to contribute to the development of an accurate system for detecting normal and abnormal heart sounds.

November 2023 Medical Image of the Month: Obstructive Uropathy Extremis

No abstract available. Manuscript truncated after 150 words. A 71-year-old veteran presented to the emergency department with two-weeks of progressive back pain radiating to the abdomen associated low-grade fever, nausea, and new lower extremity edema. The family reported confusion. His medical history was significant for chronic prostatitis and low-grade prostate cancer on biopsy that was lost to follow-up eleven years ago. His only reported medications were aspirin 81 mg daily and naproxen 500mg up to four times a day for his pain. Vitals were significant for a temperature of 36.1 C, initial blood pressure of 201/74, heart rate of 128/min, respirations at 18/min with a saturation of 97% on 2L NC. Physical exam demonstrated no difference in blood pressures between arms. No abnormal heart sounds. Clear breath sounds to auscultation bilaterally. Flank tenderness to percussion and significant abdominal tenderness over the epigastric and suprapubic region with 3+ pitting edema of the bilateral lower extremities. Screening labs were notable …

MCHeart: Multi-Channel-Based Heart Signal Processing Scheme for Heart Noise Detection Using Deep Learning.

In this study, we constructed a model to predict abnormal cardiac sounds using a diverse set of auscultation data collected from various auscultation positions. Abnormal heart sounds were identified by extracting features such as peak intervals and noise characteristics during systole and diastole. Instead of using raw signal data, we transformed them into log-mel 2D spectrograms, which were employed as input variables for the CNN model. The advancement of our model involves integrating a deep learning architecture with feature extraction techniques based on existing knowledge of cardiac data. Specifically, we propose a multi-channel-based heart signal processing (MCHeart) scheme, which incorporates our proposed features into the deep learning model. Additionally, we introduce the ReLCNN model by applying residual blocks and MHA mechanisms to the LCNN architecture. By adding murmur features with a smoothing function and training the ReLCNN model, the weighted accuracy of the model increased from 79.6% to 83.6%, showing a performance improvement of approximately 4% point compared to the LCNN baseline model.

Physician Assistant Student Attitudes Toward Gamification of Cardiac Auscultation Instruction.

Current physician assistant (PA) learners have a clear preference for interactive learning that is vibrantly present in new media technologies. At present, there is a paucity of research regarding use or acceptability of gamification in PA education. The purpose of this study was to examine PA students' experience with, attitudes toward, and outcomes of a gamified cardiac auscultation curriculum. Faculty at one institution designed an interactive Mobile App Cardiac Auscultation Curriculum (MACAC). The MACAC incorporates independent and group learning using the Littmann Learning mobile app. Author-created surveys as well as knowledge and auscultation assessment tools were delivered to all students. Most of the students recommended the use of the app for future cohorts and reported confidence to accurately identify normal and abnormal heart sounds. Knowledge and auscultation assessment scores demonstrated proficiency in identification of normal and abnormal heart sounds. Gamification research is important because blended learning that incorporates new media technologies with traditional approaches can help overcome the limitations of passive learning environments. This study provides evidence that the use of a mobile app can be an effective and innovative method to teach cardiac auscultation to the 21st century PA learners.

Detection of valvular heart diseases combining orthogonal non-negative matrix factorization and convolutional neural networks in PCG signals

Background and objective:Valvular heart disease (VHD) is associated with elevated mortality rates. Although transthoracic echocardiography (TTE) is the gold standard detection tool, phonocardiography (PCG) could be an alternative as it is a cost-effective and noninvasive method for cardiac auscultation. Many researchers have dedicated their efforts to improving the decision-making process and developing robust and precise approaches to assist physicians in providing reliable diagnoses of VHD. Methods:This research proposes a novel approach for the detection of anomalous valvular heart sounds from PCG signals. The proposed approach combines orthogonal non-negative matrix factorization (ONMF) and convolutional neural network (CNN) architectures in a three-stage cascade. The aim of the proposal is to improve the learning process by identifying the optimal ONMF temporal or spectral patterns for accurate detection. In the first stage, the time–frequency representation of the input PCG signal is computed. Next, band-pass filtering is performed to locate the spectral range that is most relevant for the presence of such cardiac abnormalities. In the second stage, the temporal and spectral cardiac structures are extracted using the ONMF approach. These structures are utilized in the third stage and fed into the CNN architecture to detect abnormal heart sounds. Results:Several state-of-the-art CNN architectures, such as LeNet5, AlexNet, ResNet50, VGG16 and GoogLeNet, have been evaluated to determine the effectiveness of using ONMF temporal features for VHD detection. The results reveal that the integration of ONMF temporal features with a CNN classifier significantly improve VHD detection. Specifically, the proposed approach achieves an accuracy improvement of approximately 45% when ONMF spectral features are used and 35% when time–frequency features from the short-time Fourier transform (STFT) spectrogram are used. Additionally, feeding ONMF temporal features into low-complexity CNN architectures yields competitive results comparable to those obtained with complex architectures. Conclusions:The temporal structure factorized by ONMF plays a critical role in distinguishing between normal heart sounds and abnormal heart sounds since the repeatability of normal heart cycles is disrupted by the presence of cardiac abnormalities. Consequently, the results highlight the importance of appropriate input data representation in the learning process of CNN models in the biomedical field of valvular heart sound detection.

Beyond Heart Murmur Detection: Automatic Murmur Grading From Phonocardiogram.

Murmurs are abnormal heart sounds, identified by experts through cardiac auscultation. The murmur grade, a quantitative measure of the murmur intensity, is strongly correlated with the patient's clinical condition. This work aims to estimate each patient's murmur grade (i.e., absent, soft, loud) from multiple auscultation location phonocardiograms (PCGs) of a large population of pediatric patients from a low-resource rural area. The Mel spectrogram representation of each PCG recording is given to an ensemble of 15 convolutional residual neural networks with channel-wise attention mechanisms to classify each PCG recording. The final murmur grade for each patient is derived based on the proposed decision rule and considering all estimated labels for available recordings. The proposed method is cross-validated on a dataset consisting of 3456 PCG recordings from 1007 patients using a stratified ten-fold cross-validation. Additionally, the method was tested on a hidden test set comprised of 1538 PCG recordings from 442 patients. The overall cross-validation performances for patient-level murmur gradings are 86.3% and 81.6% in terms of the unweighted average of sensitivities and F1-scores, respectively. The sensitivities (and F1-scores) for absent, soft, and loud murmurs are 90.7% (93.6%), 75.8% (66.8%), and 92.3% (84.2%), respectively. On the test set, the algorithm achieves an unweighted average of sensitivities of 80.4% and an F1-score of 75.8%. This study provides a potential approach for algorithmic pre-screening in low-resource settings with relatively high expert screening costs. The proposed method represents a significant step beyond detection of murmurs, providing characterization of intensity, which may provide an enhanced classification of clinical outcomes.

Lightweight network coupled with heart sounds to evaluate exercise-induced cardiac fatigue

Exercise-induced cardiac fatigue (EICF), namely a transient decline in ventricular systolic/diastolic function, usually happens in arduous endurance athletes, which increases the incidence of heart diseases. This paper proposes a computer-aided heart sound analysis method to evaluate EICF based on myocardial systolic/diastolic force, a crucial indicator for EICF. First, resting and post-exercise heart sounds were collected as dataset, and a lightweight channel grouping attention network (CGAN) was proposed to recognize abnormal heart sounds. Then, according to the data characteristic, an asymmetric depthwise separable convolution (ADSC) block was developed to extract features. The ADSC block designed based on channel grouping which can effectively reduce parameters. Finally, a channel attention module was developed to focus on kernel data features, and improved classification accuracy with negligible parameters. The CGAN achieved superior classification performance with fewer parameters, and the accuracy, precision, and recall were 96.91%, 97.01%, and 96.44% respectively. Experimental results show that our CGAN-aided heart sound analysis provides an accurate, objective, and affordable alternative to the evaluation of EICF.

Less is More: A Novel Feature Extraction Method for Heart Sound Classification via Fractal Transformation.

Cardiovascular diseases (CVDs) are the leading cause of death globally. Heart sound signal analysis plays an important role in clinical detection and physical examination of CVDs. In recent years, auxiliary diagnosis technology of CVDs based on the detection of heart sound signals has become a research hotspot. The detection of abnormal heart sounds can provide important clinical information to help doctors diagnose and treat heart disease. We propose a new set of fractal features - fractal dimension (FD) - as the representation for classification and a Support Vector Machine (SVM) as the classification model. The whole process of the method includes cutting heart sounds, feature extraction, and classification of abnormal heart sounds. We compare the classification results of the heart sound waveform (time domain) and the spectrum (frequency domain) based on fractal features. Finally, according to the better classification results, we choose the fractal features that are most conducive for classification to obtain better classification performance. The features we propose outperform the widely used features significantly (p < .05 by one-tailed z-test) with a much lower dimension.Clinical relevance-The heart sound classification model based on fractal provides a new time-frequency analysis method for heart sound signals. A new effective mechanism is proposed to explore the relationship between the heart sound acoustic properties and the pathology of CVDs. As a non-invasive diagnostic method, this work could supply an idea for the preliminary screening of cardiac abnormalities through heart sounds.

Cardiac screening in school children: Combining auscultation and electrocardiography with a crowdsourcing model

School-based cardiac screening is useful for identifying children and adolescents with a high risk of sudden cardiac death. However, because of challenges associated with cost, distance, and human resources, cardiac screening is not widely implemented, especially in rural areas with limited medical resources. This study aims to establish a cloud-based system suitable for mass cardiac screening of schoolchildren in rural areas with limited medical resources. Students from three schools were included. They or their guardians completed a simple questionnaire, administered in paper or electronic form. Heart sounds were recorded using an electronic stethoscope. Twelve-lead electrocardiograms (ECGs) were recorded and digitalized. The signals were transmitted through Bluetooth to a tablet computer and then uploaded to a cloud server over Wi-Fi. Crowdsourced pediatric cardiologists reviewed those data from a web-based platform and provided remote consultation. In cases in which abnormal heart sounds or ECGs were noted, the students were referred to the hospital for further evaluation. A total of 1004 students were enrolled in this study. Of the 138 students referred, 62 were diagnosed as having an abnormal heart condition and most had previously been undiagnosed. The interrater agreeability was high. An innovative strategy combining a cloud-based cardiac screening system with remote consultation by crowdsourced experts was established. This system allows pediatric cardiologists to provide consultation and make reliable diagnoses. Combined with crowdsourcing, the system constitutes a viable approach for mass cardiac screening in children and adolescents living in rural areas with insufficient medical resources.

Classification of Normal and Abnormal Heart Sounds Using Empirical Mode Decomposition and First Order Statistic

Analysis of heart sound signals for automatic segmentation and classification has revealed in recent decades that it has the potential to detect pathology accurately in clinical applications. Various audio signal processing techniques have been used to reduce the subjectivity of heart sound analysis. This study aims to classify normal and abnormal heart sound signals. The feature extraction process was optimized by EMD and calculated using five first-order statistical parameters: mean, variance, kurtosis, skewness, and entropy. The classification system is optimized with a mutual information algorithm to select traits that can significantly improve system performance. In addition, the selection of the optimal system configuration also includes the k-fold cross-validation and kNN methods with k values ​​and the proper distance type. Based on the test results, the highest accuracy of 98.2% was obtained when the value of k = 1 and the type of cosine distance on kNN with a five-fold cross-validation system evaluation model.

Exploring the impact of noise and degradations on heart sound classification models

The development of data-driven heart sound classification models has been an active area of research in recent years. To develop such data-driven models in the first place, heart sound signals need to be captured using a signal acquisition device. However, it is almost impossible to capture noise-free heart sound signals due to the presence of internal and external noises in most situations. Such noises and degradations in heart sound signals can potentially reduce the accuracy of data-driven classification models. Although different techniques have been proposed in the literature to address the noise issue, how and to what extent different noise and degradations in heart sound signals impact the accuracy of data-driven classification models remains unexplored. To answer this question, we produced a synthetic heart sound dataset including normal and abnormal heart sounds contaminated with a variety of noise and degradations. We used this dataset to investigate the impact of noise and degradation in heart sound recordings on the performance of different classification models. The results show different noises and degradations affect the performance of heart sound classification models to a different extent; some are more problematic for classification models, and others are less destructive. Comparing the findings of this study with the results of a survey we previously carried out with a group of clinicians shows noise and degradations that are more detrimental to classification models are also more disruptive to accurate auscultation. The findings of this study can be leveraged to develop targeted heart sound quality enhancement approaches — which adapt the type and aggressiveness of quality enhancement based on the characteristics of noise and degradation in heart sound signals.

Basic Research on the Development of an Automatic Heart Sound Diagnosis System - Analysis of Heart Sounds for Learning Policy and Experiment for the Prototype of the Auscultation Part -

A design policy was established for a specific data flow and learning method for the automatic heart-sound diagnosis system under development. The production of each part becomes possible, and auscultation and learning begin. It can be used over clothes as long as it is applied well to the skin surface of the chest. It would be nice to be able to set multiple auscultation positions, but there is a limit to what ordinary people can be asked to do, so this should be considered while having AI learn. We analyzed normal heart sounds to explore learning strategies. Sounds I and II are considered to be important anchor information sources for identifying other heart sounds. Abnormal heart sounds may not be heard at every beat and the rhythm may be abnormal. AI refers to the multiple beats of heart sounds during auscultation. Heartbeat analysis is a multidimensional information analysis related to time and space, and heart sounds are factored if normal and abnormal heart sounds can be organized based on the score. For pitch that tends to depend on individuals and devices, a relative discussion would be more appropriate.

Development of deep-learning models for a hybrid simulation of auscultation training on standard patients using an ECG-based virtual pathology stethoscope

Cardiac auscultation (CA), the act of listening to the heart’s sound, is a critical skill that provides valuable information for identifying serious heart diseases. Proficiency in cardiac auscultation requires repeated stethoscope practice and experience in identifying abnormal or irregular cardiac rhythms. However, nowadays, most hospital admissions are short and intensely focused, with fewer opportunities for medical trainees to learn and practice bedside examination skills. It is common practice in many institutions to incorporate standardized patients (SPs) into CA training because these actors are able to represent the patient and convey the symptoms. However, SPs are typically healthy individuals, limiting the kinds of abnormalities that students can hear. In this work, we develop a novel real-time simulation-based method for virtual pathology stethoscope (VPS) detection. The VPS system uses augmented reality (AR) to teach medical students how to perform cardiac examinations by listening to abnormal heart sounds in SPs who are otherwise healthy. A digital stethoscope with two electrodes on the chest piece collects electrocardiogram (ECG) signal data sets from SPs at the four primary auscultation sites. Next, different deep-learning methods are evaluated for classifying the location of the stethoscope by taking advantage of subtle differences in the ECG signals. This study would significantly extend the simulation capabilities of SPs by allowing medical students and trainees to perform realistic CA and hear CA in a clinical environment.