Classification Of Heart Murmurs Research Articles

Heart Murmur Classification Using a Capsule Neural Network.

The healthcare industry has made significant progress in the diagnosis of heart conditions due to the use of intelligent detection systems such as electrocardiograms, cardiac ultrasounds, and abnormal sound diagnostics that use artificial intelligence (AI) technology, such as convolutional neural networks (CNNs). Over the past few decades, methods for automated segmentation and classification of heart sounds have been widely studied. In many cases, both experimental and clinical data require electrocardiography (ECG)-labeled phonocardiograms (PCGs) or several feature extraction techniques from the mel-scale frequency cepstral coefficient (MFCC) spectrum of heart sounds to achieve better identification results with AI methods. Without good feature extraction techniques, the CNN may face challenges in classifying the MFCC spectrum of heart sounds. To overcome these limitations, we propose a capsule neural network (CapsNet), which can utilize iterative dynamic routing methods to obtain good combinations for layers in the translational equivariance of MFCC spectrum features, thereby improving the prediction accuracy of heart murmur classification. The 2016 PhysioNet heart sound database was used for training and validating the prediction performance of CapsNet and other CNNs. Then, we collected our own dataset of clinical auscultation scenarios for fine-tuning hyperparameters and testing results. CapsNet demonstrated its feasibility by achieving validation accuracies of 90.29% and 91.67% on the test dataset.

Heart Murmur Classification in Phonocardiogram Representations Using Convolutional Neural Networks

Heart murmurs are sounds made by rapid blood flow in the heart. Abnormal heart murmurs can be a sign of serious heart conditions such as arrhythmia and cardiovascular diseases. Therefore, heart murmur classification is crucial for early detection of such conditions. To this end, we study the heart murmur classification problem training selected convolutional neural network (CNN) models (such as VGGNet and ResNet) using various signal representations (such as spectrogram, mel-frequency cepstral coefficient (MFCC), and shorttime Fourier transform (STFT)) of the phonocardiograms in the public PASCAL CHSC dataset. Our preliminary results show that ResNet outperforms VGGNet across all metrics and representations, consistent with the recent published works we can find in literature. Unlike some of these works, however, we see MFCC and STFT in general more effective with higher test accuracies than spectrogram across all CNN models. Looking forward, we propose to study other effective models (such as InceptionV3 and Vision Transformer) to predict heart murmur conditions in phonocardiogram representations including spectrogram, MFCC and STFT, as well as others like Wigner Ville distribution.

Extraction and assessment of diagnosis-relevant features for heart murmur classification.

This paper presents a heart murmur detection and multi-class classification approach via machine learning. We extracted heart sound and murmur features that are of diagnostic importance and developed additional 16 features that are not perceivable by human ears but are valuable to improve murmur classification accuracy. We examined and compared the classification performance of supervised machine learning with k-nearest neighbor (KNN) and support vector machine (SVM) algorithms. We put together a test repertoire having more than 450 heart sound and murmur episodes to evaluate the performance of murmur classification using cross-validation of 80-20 and 90-10 splits. As clearly demonstrated in our evaluation, the specific set of features chosen in our study resulted in accurate classification consistently exceeding 90% for both classifiers.

Embedded platform based heart murmur classification using deep learning approach

Ubiquitous Perturbations in cardiac auscultation properties, cardiovascular diseases (CVDs) are widely recognized. In the auscultation procedure, the appearance of pathological cardiac murmurs is linked to heart disorders. A noble automated detection system using 1-D Convolutional Neural Network (CNN) for the detection of pathological heart murmurs is proposed in this study, which removes the difficult task of extracting and selecting features. It directly acts on the phonocardiogram (PCG) signals. The fundamental purpose of this research is to develop a classification model for consistent recognition of cardiac murmurs when the data-set is imbalanced. In view of this, the proposed study for the imbalanced data-set incorporates the Adaptive Synthetic (ADASYN) approach to generate synthetic data for the minority class. The outcome analysis illustrates the positive result in the identification of heart murmurs on both balanced and imbalanced data-sets. Therefore, the developed deep learning model will learn better from the minority class and classify heart murmurs accurately.

Narrative review of the role of artificial intelligence to improve aortic valve disease management.

Valvular heart disease (VHD) is a chronic progressive condition with an increasing prevalence in the Western world due to aging populations. VHD is often diagnosed at a late stage when patients are symptomatic and the outcomes of therapy, including valve replacement, may be sub-optimal due the development of secondary complications, including left ventricular (LV) dysfunction. The clinical application of artificial intelligence (AI), including machine learning (ML), has promise in supporting not only early and more timely diagnosis, but also hastening patient referral and ensuring optimal treatment of VHD. As physician auscultation lacks accuracy in diagnosis of significant VHD, computer-aided auscultation (CAA) with the help of a commercially available digital stethoscopes improves the detection and classification of heart murmurs. Although used little in current clinical practice, CAA can screen large populations at low cost with high accuracy for VHD and faciliate appropriate patient referral. Echocardiography remains the next step in assessment and planning management and AI is delivering major changes in speeding training, improving image quality by pattern recognition and image sorting, as well as automated measurement of multiple variables, thereby improving accuracy. Furthermore, AI then has the potential to hasten patient disposal, by automated alerts for red-flag findings, as well as decision support in dealing with results. In management, there is great potential in ML-enabled tools to support comprehensive disease monitoring and individualized treatment decisions. Using data from multiple sources, including demographic and clinical risk data to image variables and electronic reports from electronic medical records, specific patient phenotypes may be identified that are associated with greater risk or modeled to the estimate trajectory of VHD progression. Finally, AI algorithms are of proven value in planning intervention, facilitating transcatheter valve replacement by automated measurements of anatomical dimensions derived from imaging data to improve valve selection, valve size and method of delivery.

The Classification of Heart Murmurs: The Identification of Significant Time Domain Features

Phonocardiogram (PCG) is a type of acoustic signal collected from the heartbeat sound. PCG signals collected in the form of wave files and collected type of heart sound with a specific period. The difficulty of the binomial class in supervised machine learning is the steps-by-steps workflow. The consideration and decision make for every part are importantly stated so that misclassification will not occur. For the heart murmurs classification, data extraction has highly cared for it as we might have fault data consisting of outside signals. Before classifying murmurs in binomial, it will involve multiple features for selection that can have a better classification of the heart murmurs. Nevertheless, since classification performance is vital to conclude the results, models are needed to compare the research's output. The main objective of this study is to classify the signal of the murmur via time-domain based EEG signals. In this study, significant time-domain features were identified to determine the best features by using different feature selection methods. It continues with the classification with different models to compete for the highest accuracy as the performance for murmur classification. A set of Michigan Heart Sound and Murmur database (MHSDB) was provided by the University of Michigan Health System with chosen signals listening with the bell of the stethoscope at the apex area by left decubitus posture of the subjects. The PCG signals are pre-processed by segmentation of three seconds, downsampling eight thousand Hz and normalized to -1, +1. Features are extracted into ten features: Root Mean Square, Variance, Standard Deviation, Maximum, Minimum, Median, Skewness, Shape Factor, Kurtosis, and Mean. Two cross-validation methods applied, such as data splitting and k-fold cross-validation, were used to measure this study's data. Chi-Square and ANOVA technique practice to identify the significant features to improve the classification performance. The classification learners are enrolled and compared by Ada Boost, Random Forest (RF) and Support Vector Machine (SVM). The datasets will separate into a ratio of 70:30 and 80:20 for training and testing data, respectively. The chi-Square selection method was the best features selection method and 80:20 data splitting with better performance than the 70:30 ratio. The best classification accuracy for the models significantly come by SVM with all the categories with 100% except 70:30 test on test data with 97.2% only.

Microscopic abnormality classification of cardiac murmurs using ANFIS and HMM.

Auscultation of heart dispenses identification of the cardiac valves. An electronic stethoscope is used for the acquisition of heart murmurs that is further classified into normal or abnormal murmurs. The process of heart sound segmentation involves discrete wavelet transform to obtain individual components of the heart signal and its separation into systole and diastole intervals. This research presents a novel scheme to develop a semi-automatic cardiac valve disorder diagnosis system. Accordingly, features are extracted using wavelet transform and spectral analysis of input signals. The proposed classification scheme is the fusion of adaptive-neuro fuzzy inference system (ANFIS) and HMM. Both classifiers are trained using the extracted features to correctly identify normal and abnormal heart murmurs. Experimental results thus achieved exhibit that proposed system furnishes promising classification accuracy with excellent specificity and sensitivity. However, the proposed system has fewer classification errors, fewer computations, and lower dimensional feature set to build an intelligent system for detection and classification of heart murmurs.

A multi-spot exploration of the topological structures of the reconstructed phase-space for the detection of cardiac murmurs.

Acoustic heart signals are generated by a turbulence effect created when the heart valves snap shut, and therefore carrying significant information of the underlying functionality of the cardiovascular system. In this paper, we present a method for heart murmur classification divided into three major steps: a) features are extracted from the heart sound; b) features are selected using a Backward Feature Selection algorithm; c) signals are classified using a K-nearest neighbor's classifier. A new set of fractal features are proposed, which are based on the distinct signatures of complexity and self-similarity registered on the normal and pathogenic cases. The experimental results show that fractal features are the most capable of describing the non-linear structure and the underlying dynamics of heart sounds among the all feature families tested. The classification results achieved for the mitral auscultation spot (88% of accuracy) are in agreement with the current state of the art methods for heart murmur classification.

A novel hybrid energy fraction and entropy-based approach for systolic heart murmurs identification

This paper presents a set of novel features of heart sound for the detection of the abnormality of heart sounds and classification of heart murmurs. The features include energy fraction of the first and the second heart sounds (S1–S2EF), energy fraction of heart murmur (HMEF), the maximum energy fraction of heart sound frequency sub-band (HSEFmax), sample entropy of the first and the second heart sounds component (S1–S2sampen) and sample entropy of heart murmur component (HMsampen). Firstly, the heart sound signals were de-noised and normalized, then decomposed by wavelet packet. The features, such as energy fraction and sample entropy were calculated from the reconstructed selective frequency components of heart sound signals. The support vector machine (SVM) was employed as a classifier to detect the abnormality of heart sound and discriminate heart murmurs. A dataset consisting of 80 normal heart sounds and 167 systolic heart murmurs samples, segmented from 40 healthy volunteers and 67 patients, were used to test and validate the proposed method. The performance of our proposed method was assessed in terms of sensitivity, specificity and accuracy. The result showed that our proposed method exhibited a satisfactory performance with a high accuracy of 97.17%, a specificity of over 98.55% and a sensitivity of over 93.48%. This suggests that the presented method can be used as an effective assistance for cardiac auscultation.

Variable Self-Optimizing Cochlear Model for Heart Murmur Detection/Classification

Accurate detection and classification of heart murmurs by auscultation is suboptimal and not always definitive. The murmur information perceived by the physician brain is the combined effect of both patient's (human) heart and the physician's ear. The information containing the murmur characterization which is retrieved by the human brain resides in the electrical signal coming out of the cochlea. For the very reasons described here, cochlea-like processing has been successfully applied to multiple speech recognition related technologies. This had not, before our prior work, been applied to human heart murmur analysis. Our prior research consisted of three steps: (1) capturing heart sounds, (2) processing the sounds using a cochlea-like filter, and then, (3) classifying each sound as being normal or a murmur using an artificial neural network (ANN). Previously in our research, we used a static cochlea-like filter model in step 2 as described above, which resulted a significant improvement in terms of accuracy of heart murmur classification. Our cochlear filter analysis helped identify information-rich frequency segments in human heart sound. We want to advance the cochlear filter model from a static to a variable frequency selective model with feedback from ANN for better optimization of the heart murmur classification. The heart sounds will be processed in ways more closely replicating the human cochlea than the static cochlear filter. A variable self optimizing cochlear filter will better reproduce the mechanism of the human cochlea in that it will contain a feedback system from ANN to cochlear processing to automatically select the most useful frequencies based upon a threshold mechanism filtering out those frequencies which do not contain significantly useful information about classification of heart murmur. The output of the sounds in the frequency range remaining (variable self-optimizing cochlear filtered sounds) may then be used by the neural network to make a final decision about murmur classification. Our hypothesis is that a variable self-optimizing cochlear filter will significantly improve the accuracy in classification of heart sounds as normal or murmur when compared to a static cochlear filter. Using this approach, we plan to develop an AI based system which will classify heart sounds with a success rate significantly better than the static cochlear filter previously developed.

Heart Murmur Detection∕Classification System Using Cochlea-Like Pre-Processing

According to the most recent report of American Heart Association (AHA), heart disease, stroke and other cardiovascular diseases continue to remain not only the no.1 killer of Americans but also a major cause of permanent disability among American workers. Recently, many research efforts have been carried out to apply artificial intelligence (AI) to auscultation based method for rigorous detection/classification of heart murmurs but accuracy rates are not always high. All of the proposed AI techniques rely on converting the heart sound to an electrical signal and processing that signal to optimize the AI for murmur detection and classification. However, all these techniques fail to recognize that the electrical signal coming out of the cochlea is very different than the electrical signal coming out of the microphone or any other electrical sensor which is commonly used for converting heart sound to electrical signal. In this research paper, we want to take a novel approach to pre-process the electrical heart sound signal before it goes to AI for murmur detection/classification by altering the electrical signal in a similar way as is done by the human cochlea before sending the signals to the brain. Our hypothesis is that cochlea like pre-processing will change the spectral contents of the heart sound signal to enhance the murmur information which can then be efficiently detected and classified by AI circuitry. Using this approach, we plan to develop an AI based system for heart murmur classification/ detection with success rate comparable to that of an expert cardiologist.

Feature Extraction for Systolic Heart Murmur Classification

Heart murmurs are often the first signs of pathological changes of the heart valves, and they are usually found during auscultation in the primary health care. Distinguishing a pathological murmur from a physiological murmur is however difficult, why an "intelligent stethoscope" with decision support abilities would be of great value. Phonocardiographic signals were acquired from 36 patients with aortic valve stenosis, mitral insufficiency or physiological murmurs, and the data were analyzed with the aim to find a suitable feature subset for automatic classification of heart murmurs. Techniques such as Shannon energy, wavelets, fractal dimensions and recurrence quantification analysis were used to extract 207 features. 157 of these features have not previously been used in heart murmur classification. A multi-domain subset consisting of 14, both old and new, features was derived using Pudil's sequential floating forward selection (SFFS) method. This subset was compared with several single domain feature sets. Using neural network classification, the selected multi-domain subset gave the best results; 86% correct classifications compared to 68% for the first runner-up. In conclusion, the derived feature set was superior to the comparative sets, and seems rather robust to noisy data.

Comparative study of clinical evaluation of heart murmurs by general pediatricians and pediatric cardiologists.

In a study to compare the clinical diagnostic skills of academic general pediatricians and academic pediatric cardiologists in the evaluation of heart murmurs, a total of 128 patients (aged 1 month to 18 years) newly referred to a university pediatric cardiology clinic were evaluated by one of three general pediatricians and one of four pediatric cardiologists. The murmurs were clinically classified as innocent, pathologic, or possibly pathologic. The classification was revised after the review of electrocardiogram (EKG) and chest radiograph (CXR), if indicated. The definitive diagnosis was ascertained by echocardiography (94 normal, 34 abnormal). The general pediatricians identified as many pathologic heart murmurs as the pediatric cardiologists (27/34 vs. 29/34), with no difference in sensitivity, 79% vs. 85% (p = 0.53). The similarity in sensitivity could be because the general pediatricians were more cautious in the classification of heart murmurs and had classified more innocent heart murmurs as pathologic than the pediatric cardiologists (13/39 vs. 3/23), 41% vs. 13% (p = 0.02). The pediatric cardiologists correctly identified more innocent murmurs than general pediatricians (52/94 vs. 72/94), with a better specificity, 55% vs. 76% (p = 0.001); however, the accuracy of prediction of innocence was similar for both groups (52/59 vs. 72/77), 88% vs. 93% (p = 0.36). The revision of diagnosis with review of EKG and CXR was more often misleading than helpful for either group. Academic general pediatricians would identify most of the pathologic murmurs and are no more likely than an academic pediatric cardiologist to misclassify a pathologic heart murmur as innocent.