A novel method for pediatric heart sound segmentation without using the ECG

Automated Identification of Patent Ductus Arteriosus Using a Computer Vision Model

OBJECTIVES

Patent ductus arteriosus (PDA) is a vascular defect common in preterm infants and often requires treatment to avoid associated long-term morbidities. Echocardiography is the primary tool used to diagnose and monitor PDA. We trained a deep learning model to identify PDA presence in relevant echocardiographic images.

METHODS

Echocardiography video clips (n = 2527) in preterm infants were reviewed by a pediatric cardiologist and those relevant to PDA diagnosis were selected and labeled (PDA present/absent/indeterminate). We trained a convolutional neural network to classify each echocardiography frame of a clip as belonging to clips with or without PDA. A novel attention mechanism that aggregated predictions for all frames in each clip to obtain a clip-level prediction by weighting relevant frames.

RESULTS

In early model iterations, we discovered training with color Doppler echocardiography clips produced the best performing classifier. For model training and validation, 1145 such clips from 66 patients (661 PDA+ clips, 484 PDA- clips) were used. Our best classifier for clip level performance obtained sensitivity of 0.80 (0.83-0.90), specificity of 0.77 (0.62-0.92) and AUC of 0.86 (0.83-0.90). Study level performance obtained sensitivity of 0.83 (0.72-0.94), specificity of 0.89 (0.79-1.0) and AUC of 0.93 (0.89-0.98).

CONCLUSIONS

Our novel deep learning model demonstrated strong performance in classifying echocardiography clips with and without PDA. Further model development and external validation are warranted. Ultimately, integration of such a classifier into auto detection software could streamline PDA imaging workflow. This work is the first step toward semi-automated, bedside detection of PDA in preterm infants.

Markov-Based Neural Networks for Heart Sound Segmentation: Using Domain Knowledge in a Principled Way

This work considers the problem of segmenting heart sounds into their fundamental components. We unify statistical and data-driven solutions by introducing Markov-based Neural Networks (MNNs), a hybrid end-to-end framework that exploits Markov models as statistical inductive biases for an Artificial Neural Network (ANN) discriminator. We show that an MNN leveraging a simple one-dimensional Convolutional ANN significantly outperforms two recent purely data-driven solutions for this task in two publicly available datasets: PhysioNet 2016 (Sensitivity: 0.947 ±0.02; Positive Predictive Value : 0.937 ±0.025) and the CirCor DigiScope 2022 (Sensitivity: 0.950 ±0.008; Positive Predictive Value: 0.943 ±0.012). We also propose a novel gradient-based unsupervised learning algorithm that effectively makes the MNN adaptive to unseen datum sampled from unknown distributions. We perform a cross dataset analysis and show that an MNN pre-trained in the CirCor DigiScope 2022 can benefit from an average improvement of 3.90% Positive Predictive Value on unseen observations from the PhysioNet 2016 dataset using this method.

Novel phonocardiography system for heartbeat detection from various locations

The paper presents evaluation of the proposed phonocardiography (PCG) measurement system designed primarily for heartbeat detection to estimate heart rate (HR). Typically, HR estimation is performed using electrocardiography (ECG) or pulse wave as one of the fundamental diagnostic methodologies for assessing cardiac function. The system includes novel both sensory part and data processing procedure, which is based on signal preprocessing using Wavelet Transform (WT) and Shannon energy computation and heart sounds classification using K-means. Due to the lack of standardization in the placement of PCG sensors, the study focuses on evaluating the signal quality obtained from 7 different sensor locations on the subject's chest and investigates which locations are most suitable for recording heart sounds. The suitability of sensor localization was examined in 27 subjects by detecting the first two heart sounds (S1, S2). The HR detection sensitivity related to reference ECG from all sensor positions reached values over 88.9 and 77.4% in detection of S1 and S2, respectively. The placement in the middle of sternum showed the higher signal quality with median of the proper S1 and S2 detection sensitivity of 98.5 and 97.5%, respectively.

Determining the incidence of heart malformations in neonates: A novel and clinically approved solution

Background

Screening for critical congenital heart defects should be performed as early as possible and is essential for saving the lives of children and reducing the incidence of undetected adult congenital heart diseases. Heart malformations remain unrecognized at birth in more than 50% of neonates at maternity hospitals. Accurate screening for congenital heart malformations is possible using a certified and internationally patented digital intelligent phonocardiography machine. This study aimed to assess the actual incidence of heart defects in neonates. A pre-evaluation of the incidence of unrecognized severe and critical congenital heart defects at birth in our well-baby nursery was also performed.

Methods

We conducted the Neonates Cardiac Monitoring Research Project (ethics approval number: IR-IUMS-FMD. REC.1398.098) at the Shahid Akbarabadi Maternity Hospital. This study was a retrospective analysis of congenital heart malformations observed after screening 840 neonates. Using a double-blind format, 840 neonates from the well-baby nursery were randomly chosen to undergo routine clinical examinations at birth and digital intelligent phonocardiogram examinations. A pediatric cardiologist performed echocardiography for each neonate classified as having abnormal heart sounds using an intelligent machine or during routine medical examinations. If the pediatric cardiologist requested a follow-up examination, then the neonate was considered to have a congenital heart malformation, and the cumulative incidence was calculated accordingly.

Results

The incidence of heart malformations in our well-baby nursery was 5%. Furthermore, 45% of heart malformations were unrecognized in neonates at birth, including one critical congenital heart defect. The intelligent machine interpreted innocent murmurs as healthy heart sound.

Conclusion

We accurately and cost-effectively screened for congenital heart malformations in all neonates in our hospital using a digital intelligent phonocardiogram. Using an intelligent machine, we successfully identified neonates with CCHD and congenital heart defects that could not be detected using standard medical examinations. The Pouya Heart machine can record and analyze sounds with a spectral power level lower than the minimum level of the human hearing threshold. Furthermore, by redesigning the study, the identification of previously unrecognized heart malformations could increase to 58%.

Artificial Intelligence in Pediatric Cardiology: A Scoping Review

The evolution of AI and data science has aided in mechanizing several aspects of medical care requiring critical thinking: diagnosis, risk stratification, and management, thus mitigating the burden of physicians and reducing the likelihood of human error. AI modalities have expanded feet to the specialty of pediatric cardiology as well. We conducted a scoping review searching the Scopus, Embase, and PubMed databases covering the recent literature between 2002-2022. We found that the use of neural networks and machine learning has significantly improved the diagnostic value of cardiac magnetic resonance imaging, echocardiograms, computer tomography scans, and electrocardiographs, thus augmenting the clinicians' diagnostic accuracy of pediatric heart diseases. The use of AI-based prediction algorithms in pediatric cardiac surgeries improves postoperative outcomes and prognosis to a great extent. Risk stratification and the prediction of treatment outcomes are feasible using the key clinical findings of each CHD with appropriate computational algorithms. Notably, AI can revolutionize prenatal prediction as well as the diagnosis of CHD using the EMR (electronic medical records) data on maternal risk factors. The use of AI in the diagnostics, risk stratification, and management of CHD in the near future is a promising possibility with current advancements in machine learning and neural networks. However, the challenges posed by the dearth of appropriate algorithms and their nascent nature, limited physician training, fear of over-mechanization, and apprehension of missing the 'human touch' limit the acceptability. Still, AI proposes to aid the clinician tomorrow with precision cardiology, paving a way for extremely efficient human-error-free health care.

Unsupervised detection and classification of heartbeats using the dissimilarity matrix in PCG signals

BACKGROUND AND OBJECTIVE

Auscultation is the first technique applied to the early diagnose of any cardiovascular disease (CVD) in rural areas and poor-resources countries because of its low cost and non-invasiveness. However, it highly depends on the physician's expertise to recognize specific heart sounds heard through the stethoscope. The analysis of phonocardiogram (PCG) signals attempts to segment each cardiac cycle into the four cardiac states (S1, systole, S2 and diastole) in order to develop automatic systems applied to an efficient and reliable detection and classification of heartbeats. In this work, we propose an unsupervised approach, based on time-frequency characteristics shown by cardiac sounds, to detect and classify heartbeats S1 and S2.

METHODS

The proposed system consists of a two-stage cascade. The first stage performs a rough heartbeat detection while the second stage refines the previous one, improving the temporal localization and also classifying the heartbeats into types S1 and S2. The first contribution is a novel approach that combines the dissimilarity matrix with the frame-level spectral divergence to locate heartbeats using the repetitiveness shown by the heart sounds and the temporal relationships between the intervals defined by the events S1/S2 and non-S1/S2 (systole and diastole). The second contribution is a verification-correction-classification process based on a sliding window that allows the preservation of the temporal structure of the cardiac cycle in order to be applied in the heart sound classification. The proposed method has been assessed using the open access databases PASCAL, CirCor DigiScope Phonocardiogram and an additional sound mixing procedure considering both Additive White Gaussian Noise (AWGN) and different kinds of clinical ambient noises from a commercial database.

RESULTS

The proposed method outperforms the detection and classification performance of other recent state-of-the-art methods. Although our proposal achieves the best average accuracy for PCG signals without cardiac abnormalities, 99.4% in heartbeat detection and 97.2% in heartbeat classification, its worst average accuracy is always above 92% for PCG signals with cardiac abnormalities, signifying an improvement in heartbeat detection/classification above 10% compared to the other state-of-the-art methods evaluated.

CONCLUSIONS

The proposed method provides the best detection/classification performance in realistic scenarios where the presence of cardiac anomalies as well as different types of clinical environmental noises are active in the PCG signal. Of note, the promising modelling of the temporal structures of the heart provided by the dissimilarity matrix together with the frame-level spectral divergence, as well as the removal of a significant number of spurious heart events and recovery of missing heart events, both corrected by the proposed verification-correction-classification algorithm, suggest that our proposal is a successful tool to be applied in heart segmentation.

The discerning ear: cardiac auscultation in the era of artificial intelligence and telemedicine

Heart murmur, a thoracic auscultatory finding of cardiovascular origin, is extremely common in childhood and can appear at any age from premature newborn to late adolescence. The objective of this review is to provide a modern examination and update of cardiac murmur auscultation in this new era of artificial intelligence (AI) and telemedicine. First, we provide a comprehensive review of the causes and differential diagnosis, clinical features, evaluation, and long-term management of paediatric heart murmurs. Next, we provide a brief history of computer-assisted auscultation and murmur analysis, along with insight into the engineering design of the digital stethoscope. We conclude with a discussion of the paradigm shifting impact of deep learning on murmur analysis, AI-assisted auscultation, and the implications of these technologies on telemedicine in paediatric cardiology. It is our hope that this article provides an updated perspective on the impact of AI on cardiac auscultation for the modern paediatric cardiologist.

Cross-Domain Transfer Learning for PCG Diagnosis Algorithm

Cardiechema is a way to reflect cardiovascular disease where the doctor uses a stethoscope to help determine the heart condition with a sound map. In this paper, phonocardiogram (PCG) is used as a diagnostic signal, and a deep learning diagnostic framework is proposed. By improving the architecture and modules, a new transfer learning and boosting architecture is mainly employed. In addition, a segmentation method is designed to improve on the existing signal segmentation methods, such as R wave to R wave interval segmentation and fixed segmentation. For the evaluation, the final diagnostic architecture achieved a sustainable performance with a public PCG database.

End-to-end heart sound segmentation using deep convolutional recurrent network

Heart sound segmentation (HSS) aims to detect the four stages (first sound, systole, second heart sound and diastole) from a heart cycle in a phonocardiogram (PCG), which is an essential step in automatic auscultation analysis. Traditional HSS methods need to manually extract the features before dealing with HSS tasks. These artificial features highly rely on extraction algorithms, which often result in poor performance due to the different operating environments. In addition, the high-dimension and frequency characteristics of audio also challenge the traditional methods in effectively addressing HSS tasks. This paper presents a novel end-to-end method based on convolutional long short-term memory (CLSTM), which directly uses audio recording as input to address HSS tasks. Particularly, the convolutional layers are designed to extract the meaningful features and perform the downsampling, and the LSTM layers are developed to conduct the sequence recognition. Both components collectively improve the robustness and adaptability in processing the HSS tasks. Furthermore, the proposed CLSTM algorithm is easily extended to other complex heart sound annotation tasks, as it does not need to extract the characteristics of corresponding tasks in advance. In addition, the proposed algorithm can also be regarded as a powerful feature extraction tool, which can be integrated into the existing models for HSS. Experimental results on real-world PCG datasets, through comparisons to peer competitors, demonstrate the outstanding performance of the proposed algorithm.

An innovative method for cardiovascular disease detection based on nonlinear geometric features and feature reduction combination

Cardiovascular is arguably the most dominant death cause in the world. Heart functionality can be measured in various ways. Heart sounds are usually inspected in these experiments as they can unveil a variety of heart related diseases. This study tackles the lack of reliable models and high training times on a publicly available dataset. The heart sound set is provided by Physionet consisting of 3153 recordings, from which five seconds were fixed to evaluate to the developed method. In this work, we propose a novel method based on feature reduction combination, using Genetic Algorithm (GA) and Principal Component Analysis (PCA). The authors present eight dominant features in heart sound classification: mean duration of systole interval, the standard deviation of diastole interval, the absolute amplitude ratio of diastole to S2, S1 to systole and S1 to diastole, zero crossings, Centroid to Centroid distance (CCdis) and mean power in the 95–295 Hz range. These reduced features are then optimized respectively with two straightforward classification algorithms weighted k-NN with a lower-dimensional feature space and Linear SVM that uses a linear combination of all features to create a robust model, acquiring up to 98.15% accuracy, holding the best stats in the heart sound classification on a largely used dataset. According to the experiments done in this study, the developed method can be further explored for real world heart sound assessments.

Temporal Convolutional Network Connected with an Anti-Arrhythmia Hidden Semi-Markov Model for Heart Sound Segmentation

Heart sound segmentation (HSS) is a critical step in heart sound processing, where it improves the interpretability of heart sound disease classification algorithms. In this study, we aimed to develop a real-time algorithm for HSS by combining the temporal convolutional network (TCN) and the hidden semi-Markov model (HSMM), and improve the performance of HSMM for heart sounds with arrhythmias. We experimented with TCN and determined the best parameters based on spectral features, envelopes, and one-dimensional CNN. However, the TCN results could contradict the natural fixed order of S1-systolic-S2-diastolic of heart sound, and thereby the Viterbi algorithm based on HSMM was connected to correct the order errors. On this basis, we improved the performance of the Viterbi algorithm when detecting heart sounds with cardiac arrhythmias by changing the distribution and weights of the state duration probabilities. The public PhysioNet Computing in Cardiology Challenge 2016 data set was employed to evaluate the performance of the proposed algorithm. The proposed algorithm achieved an F1 score of 97.02%, and this result was comparable with the current state-of-the-art segmentation algorithms. In addition, the proposed enhanced Viterbi algorithm for HSMM corrected 30 out of 30 arrhythmia errors after checking one by one in the dataset.

Intelligent Diagnosis of Heart Murmurs in Children with Congenital Heart Disease

Heart auscultation is a convenient tool for early diagnosis of heart diseases and is being developed to be an intelligent tool used in online medicine. Currently, there are few studies on intelligent diagnosis of pediatric murmurs due to congenital heart disease (CHD). The purpose of the study was to develop a method of intelligent diagnosis of pediatric CHD murmurs. Phonocardiogram (PCG) signals of 86 children were recorded with 24 children having normal heart sounds and 62 children having CHD murmurs. A segmentation method based on the discrete wavelet transform combined with Hadamard product was implemented to locate the first and the second heart sounds from the PCG signal. Ten features specific to CHD murmurs were extracted as the input of classifier after segmentation. Eighty-six artificial neural network classifiers were composed into a classification system to identify CHD murmurs. The accuracy, sensitivity, and specificity of diagnosis for heart murmurs were 93%, 93.5%, and 91.7%, respectively. In conclusion, a method of intelligent diagnosis of pediatric CHD murmurs is developed successfully and can be used for online screening of CHD in children.

Is there an Optimal Localization of Cardio-microphone Sensors for Phonocardiogram Analysis?

Heart auscultation is one of the most useful medical diagnostic tools for getting valuable information of heart valves and heart hemodynamics functions. However, the information acquired by a traditional stethoscope can be inaccurate and insufficient. Phonocardiogram (PCG) was developed to improve accuracy through visual inspection and analysis. Digitally processed, PCG can then be analyzed by automated heart sound analysis systems. But there is no standardization for PCG data acquisition unlike electrocardiogram (ECG). This study aims at analyzing the influence of cardiomicrophone localization on the chest for the study of cardiac sounds S1 and S2. For that purpose, simultaneous acquisitions of 12 PCG signals with one ECG signal were realized and a comparative analysis of delays between R waves of ECG and detected S1 and S2 sounds was conducted. Results show that there are significant differences between R-S1 (or R-S2) intervals obtained from different areas of sensor placement on the chest. For future works on PCG, studies dealing with the analysis of heart sounds or proposing new heart sounds detection algorithms may pay attention to the location and attachment of PCG sensors.

Heart Sound Segmentation Using Bidirectional LSTMs With Attention

OBJECTIVE

This paper proposes a novel framework for the segmentation of phonocardiogram (PCG) signals into heart states, exploiting the temporal evolution of the PCG as well as considering the salient information that it provides for the detection of the heart state.

METHODS

We propose the use of recurrent neural networks and exploit recent advancements in attention based learning to segment the PCG signal. This allows the network to identify the most salient aspects of the signal and disregard uninformative information.

RESULTS

The proposed method attains state-of-the-art performance on multiple benchmarks including both human and animal heart recordings. Furthermore, we empirically analyse different feature combinations including envelop features, wavelet and Mel Frequency Cepstral Coefficients (MFCC), and provide quantitative measurements that explore the importance of different features in the proposed approach.

CONCLUSION

We demonstrate that a recurrent neural network coupled with attention mechanisms can effectively learn from irregular and noisy PCG recordings. Our analysis of different feature combinations shows that MFCC features and their derivatives offer the best performance compared to classical wavelet and envelop features.

SIGNIFICANCE

Heart sound segmentation is a crucial pre-processing step for many diagnostic applications. The proposed method provides a cost effective alternative to labour extensive manual segmentation, and provides a more accurate segmentation than existing methods. As such, it can improve the performance of further analysis including the detection of murmurs and ejection clicks. The proposed method is also applicable for detection and segmentation of other one dimensional biomedical signals.

A Markov-Switching Model Approach to Heart Sound Segmentation and Classification

OBJECTIVE

We consider challenges in accurate segmentation of heart sound signals recorded under noisy clinical environments for subsequent classification of pathological events. Existing state-of-the-art solutions to heart sound segmentation use probabilistic models such as hidden Markov models (HMMs), which, however, are limited by its observation independence assumption and rely on pre-extraction of noise-robust features.

METHODS

We propose a Markov-switching autoregressive (MSAR) process to model the raw heart sound signals directly, which allows efficient segmentation of the cyclical heart sound states according to the distinct dependence structure in each state. To enhance robustness, we extend the MSAR model to a switching linear dynamic system (SLDS) that jointly model both the switching AR dynamics of underlying heart sound signals and the noise effects. We introduce a novel algorithm via fusion of switching Kalman filter and the duration-dependent Viterbi algorithm, which incorporates the duration of heart sound states to improve state decoding.

RESULTS

Evaluated on Physionet/CinC Challenge 2016 dataset, the proposed MSAR-SLDS approach significantly outperforms the hidden semi-Markov model (HSMM) in heart sound segmentation based on raw signals and comparable to a feature-based HSMM. The segmented labels were then used to train Gaussian-mixture HMM classifier for identification of abnormal beats, achieving high average precision of 86.1% on the same dataset including very noisy recordings.

CONCLUSION

The proposed approach shows noticeable performance in heart sound segmentation and classification on a large noisy dataset.

SIGNIFICANCE

It is potentially useful in developing automated heart monitoring systems for pre-screening of heart pathologies.

Echocardiographic Evaluation in Neonates with Heart Murmurs

Auscultation of the heart is one of the most important tools of physical examination in neonates, which is very helpful in the diagnosis or rejection of congenital heart diseases (CHDs). The aim of this study is to investigate the causes of heart murmur in neonates hospitalized in neonatal unit in Imam Reza Hospital, Kermanshah, Iran. This cross-sectional study was conducted within the period of 18 months on neonates hospitalized in neonatal unit whose heart murmur, diagnosed through examination, and echocardiography was conducted to investigate the existence of CHD. In 62.1% of the cases, murmuring was auscultated within the first week after birth. Among 172 neonates with heart murmur, 25 subjects (14.5%) had normal echocardiogram, and others had abnormal echocardiogram in which ventricular septal defect (31.4%) was the most common CHD. Patent ductus arteriosus (23.3%) was the second common CHD found. Other defects were atrial septal defect, persistent foramen ovale, pulmonary stenosis, and tricuspid regurgitation. Among the neonates with CHD, 55 subjects were females and 92 cases were males. There was CHD history in previous siblings in 10 cases. In 73 neonates with CHD, the mother had not taken folic acid in the first trimester of pregnancy. A total of 14.5% of the hospitalized neonates suffered from innocent murmur, and the rest with heart murmur had abnormal echocardiography and suffered from CHD (85.5%). Heart murmur in neonates could be a symptom of CHD, and timely echocardiography is very important in diagnosing the type of disease.

Deep Convolutional Neural Networks for Heart Sound Segmentation

This paper studies the use of deep convolutional neural networks to segment heart sounds into their main components. The proposed methods are based on the adoption of a deep convolutional neural network architecture, which is inspired by similar approaches used for image segmentation. Different temporal modeling schemes are applied to the output of the proposed neural network, which induce the output state sequence to be consistent with the natural sequence of states within a heart sound signal (S1, systole, S2, diastole). In particular, convolutional neural networks are used in conjunction with underlying hidden Markov models and hidden semi-Markov models to infer emission distributions. The proposed approaches are tested on heart sound signals from the publicly available PhysioNet dataset, and they are shown to outperform current state-of-the-art segmentation methods by achieving an average sensitivity of 93.9% and an average positive predictive value of 94% in detecting S1 and S2 sounds.

A Deep Machine Learning Method for Classifying Cyclic Time Series of Biological Signals Using Time-Growing Neural Network

This paper presents a novel method for learning the cyclic contents of stochastic time series: the deep time-growing neural network (DTGNN). The DTGNN combines supervised and unsupervised methods in different levels of learning for an enhanced performance. It is employed by a multiscale learning structure to classify cyclic time series (CTS), in which the dynamic contents of the time series are preserved in an efficient manner. This paper suggests a systematic procedure for finding the design parameter of the classification method for a one-versus-multiple class application. A novel validation method is also suggested for evaluating the structural risk, both in a quantitative and a qualitative manner. The effect of the DTGNN on the performance of the classifier is statistically validated through the repeated random subsampling using different sets of CTS, from different medical applications. The validation involves four medical databases, comprised of 108 recordings of the electroencephalogram signal, 90 recordings of the electromyogram signal, 130 recordings of the heart sound signal, and 50 recordings of the respiratory sound signal. Results of the statistical validations show that the DTGNN significantly improves the performance of the classification and also exhibits an optimal structural risk.

An automatic segmentation method for heart sounds

BACKGROUND

There are two major challenges in automated heart sound analysis: segmentation and classification. An efficient segmentation is capable of providing valuable diagnostic information of patients. In addition, it is crucial for some feature-extraction based classification methods. Therefore, the segmentation of heart sound is of significant value.

METHODS

This paper presents an automatic heart sound segmentation method that combines the time-domain analysis, frequency-domain analysis and time-frequency-domain analysis. Employing this method, the boundaries of heart sound components are first located, and the components are then recognized. Finally, the heart sounds are divided into several segments on the basis of the results of boundary localization and component identification.

RESULTS

In order to evaluate the performance of the proposed method, quantitative experiments are performed on an authoritative heart sound database. The experimental results show that the boundary localization has a sensitivity (Se) of 100%, a positive predictive value (PPV) of 99.3% and an accuracy (Acc) of 99.93%. Moreover, the Se, PPV and Acc of component identification reach 98.63, 99.86 and 98.49%, respectively.

CONCLUSION

The proposed method shows reliable performance on the segmentation of heart sounds. Compared with previous works, this method can be applied to not only normal heart sounds, but also the sounds with S3, S4 and murmurs, thus greatly increasing the applied range.

Heart Sound Segmentation-An Event Detection Approach Using Deep Recurrent Neural Networks

OBJECTIVE

In this paper, we accurately detect the state-sequence first heart sound (S1)-systole-second heart sound (S2)-diastole, i.e., the positions of S1 and S2, in heart sound recordings. We propose an event detection approach without explicitly incorporating a priori information of the state duration. This renders it also applicable to recordings with cardiac arrhythmia and extendable to the detection of extra heart sounds (third and fourth heart sound), heart murmurs, as well as other acoustic events.

METHODS

We use data from the 2016 PhysioNet/CinC Challenge, containing heart sound recordings and annotations of the heart sound states. From the recordings, we extract spectral and envelope features and investigate the performance of different deep recurrent neural network (DRNN) architectures to detect the state sequence. We use virtual adversarial training, dropout, and data augmentation for regularization.

RESULTS

We compare our results with the state-of-the-art method and achieve an average score for the four events of the state sequence of ${\bf F}_{1}\approx 96$% on an independent test set.

CONCLUSION

Our approach shows state-of-the-art performance carefully evaluated on the 2016 PhysioNet/CinC Challenge dataset.

SIGNIFICANCE

In this work, we introduce a new methodology for the segmentation of heart sounds, suggesting an event detection approach with DRNNs using spectral or envelope features.

Performance of an open-source heart sound segmentation algorithm on eight independent databases

OBJECTIVE

Heart sound segmentation is a prerequisite step for the automatic analysis of heart sound signals, facilitating the subsequent identification and classification of pathological events. Recently, hidden Markov model-based algorithms have received increased interest due to their robustness in processing noisy recordings. In this study we aim to evaluate the performance of the recently published logistic regression based hidden semi-Markov model (HSMM) heart sound segmentation method, by using a wider variety of independently acquired data of varying quality.

APPROACH

Firstly, we constructed a systematic evaluation scheme based on a new collection of heart sound databases, which we assembled for the PhysioNet/CinC Challenge 2016. This collection includes a total of more than 120 000 s of heart sounds recorded from 1297 subjects (including both healthy subjects and cardiovascular patients) and comprises eight independent heart sound databases sourced from multiple independent research groups around the world. Then, the HSMM-based segmentation method was evaluated using the assembled eight databases. The common evaluation metrics of sensitivity, specificity, accuracy, as well as the [Formula: see text] measure were used. In addition, the effect of varying the tolerance window for determining a correct segmentation was evaluated.

MAIN RESULTS

The results confirm the high accuracy of the HSMM-based algorithm on a separate test dataset comprised of 102 306 heart sounds. An average [Formula: see text] score of 98.5% for segmenting S1 and systole intervals and 97.2% for segmenting S2 and diastole intervals were observed. The [Formula: see text] score was shown to increases with an increases in the tolerance window size, as expected.

SIGNIFICANCE

The high segmentation accuracy of the HSMM-based algorithm on a large database confirmed the algorithm's effectiveness. The described evaluation framework, combined with the largest collection of open access heart sound data, provides essential resources for evaluators who need to test their algorithms with realistic data and share reproducible results.

Enhancement of lung sounds based on empirical mode decomposition and Fourier transform algorithm

BACKGROUND AND OBJECTIVE

There is always heart sound (HS) signal interfering during the recording of lung sound (LS) signals. This obscures the features of LS signals and creates confusion on pathological states, if any, of the lungs. In this work, a new method is proposed for reduction of heart sound interference which is based on empirical mode decomposition (EMD) technique and prediction algorithm.

METHOD

In this approach, first the mixed signal is split into several components in terms of intrinsic mode functions (IMFs). Thereafter, HS-included segments are localized and removed from them. The missing values of the gap thus produced, is predicted by a new Fast Fourier Transform (FFT) based prediction algorithm and the time domain LS signal is reconstructed by taking an inverse FFT of the estimated missing values.

RESULTS

The experiments have been conducted on simulated and recorded HS corrupted LS signals at three different flow rates and various SNR levels. The performance of the proposed method is evaluated by qualitative and quantitative analysis of the results.

CONCLUSIONS

It is found that the proposed method is superior to the baseline method in terms of quantitative and qualitative measurement. The developed method gives better results compared to baseline method for different SNR levels. Our method gives cross correlation index (CCI) of 0.9488, signal to deviation ratio (SDR) of 9.8262, and normalized maximum amplitude error (NMAE) of 26.94 for 0 dB SNR value.

An open access database for the evaluation of heart sound algorithms

In the past few decades, analysis of heart sound signals (i.e. the phonocardiogram or PCG), especially for automated heart sound segmentation and classification, has been widely studied and has been reported to have the potential value to detect pathology accurately in clinical applications. However, comparative analyses of algorithms in the literature have been hindered by the lack of high-quality, rigorously validated, and standardized open databases of heart sound recordings. This paper describes a public heart sound database, assembled for an international competition, the PhysioNet/Computing in Cardiology (CinC) Challenge 2016. The archive comprises nine different heart sound databases sourced from multiple research groups around the world. It includes 2435 heart sound recordings in total collected from 1297 healthy subjects and patients with a variety of conditions, including heart valve disease and coronary artery disease. The recordings were collected from a variety of clinical or nonclinical (such as in-home visits) environments and equipment. The length of recording varied from several seconds to several minutes. This article reports detailed information about the subjects/patients including demographics (number, age, gender), recordings (number, location, state and time length), associated synchronously recorded signals, sampling frequency and sensor type used. We also provide a brief summary of the commonly used heart sound segmentation and classification methods, including open source code provided concurrently for the Challenge. A description of the PhysioNet/CinC Challenge 2016, including the main aims, the training and test sets, the hand corrected annotations for different heart sound states, the scoring mechanism, and associated open source code are provided. In addition, several potential benefits from the public heart sound database are discussed.

Automated Identification of Innocent Still's Murmur in Children

OBJECTIVE

Still's murmur is the most common innocent heart murmur in children. It is also the most commonly misdiagnosed murmur, resulting in a high number of unnecessary referrals to pediatric cardiologist. The purpose of this study was to develop a computer algorithm for automated identification of Still's murmur that may help reduce unnecessary referrals.

METHODS

We first developed an accurate segmentation algorithm to locate the first and the second heart sounds. Once these sounds were identified, we extracted signal features specific to Still's murmur. Subsequently, machine learning-based classifiers, artificial neural network and support vector machine, were used to identify Still's murmur.

RESULTS

We evaluated our classifiers using the jackknife method using 87 Still's murmurs and 170 non-Still's murmurs. Our algorithm identified Still's murmur accurately with 84-93% sensitivity and 91-99% specificity.

CONCLUSION

We have achieved accurate automated identification of Still's murmur while minimizing false positives. The performance of our algorithm is comparable to the rate of murmur identification by auscultation by pediatric cardiologists.

SIGNIFICANCE

To our knowledge, our solution is the first murmur classifier that focuses singularly on Still's murmur. Following further refinement and testing, the presented algorithm could reduce the number of children with Still's murmur referred unnecessarily to pediatric cardiologists.

Detection of Heart Sounds in Children with and without Pulmonary Arterial Hypertension--Daubechies Wavelets Approach

Background

Automatic detection of the 1^st (S1) and 2^nd (S2) heart sounds is difficult, and existing algorithms are imprecise. We sought to develop a wavelet-based algorithm for the detection of S1 and S2 in children with and without pulmonary arterial hypertension (PAH).

Method

Heart sounds were recorded at the second left intercostal space and the cardiac apex with a digital stethoscope simultaneously with pulmonary arterial pressure (PAP). We developed a Daubechies wavelet algorithm for the automatic detection of S1 and S2 using the wavelet coefficient ‘D₆’ based on power spectral analysis. We compared our algorithm with four other Daubechies wavelet-based algorithms published by Liang, Kumar, Wang, and Zhong. We annotated S1 and S2 from an audiovisual examination of the phonocardiographic tracing by two trained cardiologists and the observation that in all subjects systole was shorter than diastole.

Results

We studied 22 subjects (9 males and 13 females, median age 6 years, range 0.25–19). Eleven subjects had a mean PAP < 25 mmHg. Eleven subjects had PAH with a mean PAP ≥ 25 mmHg. All subjects had a pulmonary artery wedge pressure ≤ 15 mmHg. The sensitivity (SE) and positive predictivity (+P) of our algorithm were 70% and 68%, respectively. In comparison, the SE and +P of Liang were 59% and 42%, Kumar 19% and 12%, Wang 50% and 45%, and Zhong 43% and 53%, respectively. Our algorithm demonstrated robustness and outperformed the other methods up to a signal-to-noise ratio (SNR) of 10 dB. For all algorithms, detection errors arose from low-amplitude peaks, fast heart rates, low signal-to-noise ratio, and fixed thresholds.

Conclusion

Our algorithm for the detection of S1 and S2 improves the performance of existing Daubechies-based algorithms and justifies the use of the wavelet coefficient ‘D₆’ through power spectral analysis. Also, the robustness despite ambient noise may improve real world clinical performance.

Pediatric heart sound segmentation using hidden Markov model

Recent advances in technology have enabled automatic cardiac auscultation using digital stethoscopes. This in turn creates the need for development of algorithms capable of automatic segmentation of heart sounds. Pediatric heart sound segmentation is a challenging task due to various confounding factors including the significant influence of respiration on children's heart sounds. The current work investigates the application of homomorphic filtering and Hidden Markov Model for the purpose of segmenting pediatric heart sounds. The efficacy of the proposed method is evaluated on the publicly available Pascal Challenge dataset and its performance is compared with those of three other existing methods. The results show that our proposed method achieves an accuracy of 92.4%±1.1% and 93.5%±1.1% in identifying the first and second heart sound components, respectively, and is superior to three other existing methods in terms of accuracy or computational complexity.

An Intelligent Phonocardiography for Automated Screening of Pediatric Heart Diseases

This paper presents a robust device for automated screening of pediatric heart diseases based on our unique processing method in murmur characterization; the Arash-Band method. The present study modifies the Arash-Band method and employs output of the modified method in conjunction with the two other original techniques to extract indicative feature vectors for the screening. The extracted feature vectors are classified by using the support vector machine method. Results show that the proposed modifications significantly enhances performance of the Arash-Band in terms of the both accuracy and sensitivity as the corresponding effect sizes are sufficiently large. The proposed algorithm has been incorporated into an Android-based tablet to constitute an intelligent phonocardiogram with the automatic screening capability. In order to obtain confidence interval of the accuracy and sensitivity, an inferable statistical test is applied on our database containing the phonocardiogram signals recorded from 263 of the referrals to a hospital. The expected value of the accuracy/sensitivity is estimated to be 87.45 % / 87.29 % with a 95 % confidence interval of (80.19 % - 92.47 %) / (76.01 % - 95.78 %) exhibiting superior performance than a pediatric cardiologist who relies on conventional or even computer-assisted auscultation.

Logistic Regression-HSMM-Based Heart Sound Segmentation

The identification of the exact positions of the first and second heart sounds within a phonocardiogram (PCG), or heart sound segmentation, is an essential step in the automatic analysis of heart sound recordings, allowing for the classification of pathological events. While threshold-based segmentation methods have shown modest success, probabilistic models, such as hidden Markov models, have recently been shown to surpass the capabilities of previous methods. Segmentation performance is further improved when a priori information about the expected duration of the states is incorporated into the model, such as in a hidden semi-Markov model (HSMM). This paper addresses the problem of the accurate segmentation of the first and second heart sound within noisy real-world PCG recordings using an HSMM, extended with the use of logistic regression for emission probability estimation. In addition, we implement a modified Viterbi algorithm for decoding the most likely sequence of states, and evaluated this method on a large dataset of 10,172 s of PCG recorded from 112 patients (including 12,181 first and 11,627 second heart sounds). The proposed method achieved an average F1 score of 95.63 ± 0.85%, while the current state of the art achieved 86.28 ± 1.55% when evaluated on unseen test recordings. The greater discrimination between states afforded using logistic regression as opposed to the previous Gaussian distribution-based emission probability estimation as well as the use of an extended Viterbi algorithm allows this method to significantly outperform the current state-of-the-art method based on a two-sided paired t-test.

A Novel Method for Screening Children with Isolated Bicuspid Aortic Valve

This paper presents a novel processing method for heart sound signal: the statistical time growing neural network (STGNN). The STGNN performs a robust classification by merging supervised and unsupervised statistical methods to overcome non-stationary behavior of the signal. By combining available preprocessing and segmentation techniques and the STGNN classifier, we build an automatic tool for screening children with isolated BAV, the congenital heart malformation which can lead to serious cardiovascular lesions. Children with BAV (22 individuals) and healthy condition (28 individuals) are subjected to the study. The performance of the STGNN is compared to that of a time growing neural network (CTGNN) and a conventional support vector (CSVM) machine, using balanced repeated random sub sampling. The average of the accuracy/sensitivity is estimated to be 87.4/86.5 for the STGNN, 81.8/83.4 for the CTGNN, and 72.9/66.8 for the CSVM. Results show that the STGNN offers better performance and provides more immunity to the background noise as compared to the CTGNN and CSVM. The method is implementable in a computer system to be employed in primary healthcare centers to improve the screening accuracy.

High Order Statistics and Time-Frequency Domain to Classify Heart Sounds for Subjects under Cardiac Stress Test

This paper considers the problem of classification of the first and the second heart sounds (S1 and S2) under cardiac stress test. The main objective is to classify these sounds without electrocardiogram (ECG) reference and without taking into consideration the systolic and the diastolic time intervals criterion which can become problematic and useless in several real life settings as severe tachycardia and tachyarrhythmia or in the case of subjects being under cardiac stress activity. First, the heart sounds are segmented by using a modified time-frequency based envelope. Then, to distinguish between the first and the second heart sounds, new features, named α_opt, β, and γ, based on high order statistics and energy concentration measures of the Stockwell transform (S-transform) are proposed in this study. A study of the variation of the high frequency content of S1 and S2 over the HR (heart rate) is also discussed. The proposed features are validated on a database that contains 2636 S1 and S2 sounds corresponding to 62 heart signals and 8 subjects under cardiac stress test collected from healthy subjects. Results and comparisons with existing methods in the literature show a large superiority for our proposed features.

Automatic moment segmentation and peak detection analysis of heart sound pattern via short-time modified Hilbert transform

This paper proposes a novel automatic method for the moment segmentation and peak detection analysis of heart sound (HS) pattern, with special attention to the characteristics of the envelopes of HS and considering the properties of the Hilbert transform (HT). The moment segmentation and peak location are accomplished in two steps. First, by applying the Viola integral waveform method in the time domain, the envelope (E(T)) of the HS signal is obtained with an emphasis on the first heart sound (S1) and the second heart sound (S2). Then, based on the characteristics of the E(T) and the properties of the HT of the convex and concave functions, a novel method, the short-time modified Hilbert transform (STMHT), is proposed to automatically locate the moment segmentation and peak points for the HS by the zero crossing points of the STMHT. A fast algorithm for calculating the STMHT of E(T) can be expressed by multiplying the E(T) by an equivalent window (W(E)). According to the range of heart beats and based on the numerical experiments and the important parameters of the STMHT, a moving window width of N=1s is validated for locating the moment segmentation and peak points for HS. The proposed moment segmentation and peak location procedure method is validated by sounds from Michigan HS database and sounds from clinical heart diseases, such as a ventricular septal defect (VSD), an aortic septal defect (ASD), Tetralogy of Fallot (TOF), rheumatic heart disease (RHD), and so on. As a result, for the sounds where S2 can be separated from S1, the average accuracies achieved for the peak of S1 (AP₁), the peak of S2 (AP₂), the moment segmentation points from S1 to S2 (AT₁₂) and the cardiac cycle (ACC) are 98.53%, 98.31% and 98.36% and 97.37%, respectively. For the sounds where S1 cannot be separated from S2, the average accuracies achieved for the peak of S1 and S2 (AP₁₂) and the cardiac cycle ACC are 100% and 96.69%.

Frequency shifting approach towards textual transcription of heartbeat sounds

Auscultation is an approach for diagnosing many cardiovascular problems. Automatic analysis of heartbeat sounds and extraction of its audio features can assist physicians towards diagnosing diseases. Textual transcription allows recording a continuous heart sound stream using a text format which can be stored in very small memory in comparison with other audio formats. In addition, a text-based data allows applying indexing and searching techniques to access to the critical events. Hence, the transcribed heartbeat sounds provides useful information to monitor the behavior of a patient for the long duration of time. This paper proposes a frequency shifting method in order to improve the performance of the transcription. The main objective of this study is to transfer the heartbeat sounds to the music domain. The proposed technique is tested with 100 samples which were recorded from different heart diseases categories. The observed results show that, the proposed shifting method significantly improves the performance of the transcription.

Signal processing of heart signals for the quantification of non-deterministic events

BACKGROUND

Heart signals represent an important way to evaluate cardiovascular function and often what is desired is to quantify the level of some signal of interest against the louder backdrop of the beating of the heart itself. An example of this type of application is the quantification of cavitation in mechanical heart valve patients.

METHODS

An algorithm is presented for the quantification of high-frequency, non-deterministic events such as cavitation from recorded signals. A closed-form mathematical analysis of the algorithm investigates its capabilities. The algorithm is implemented on real heart signals to investigate usability and implementation issues. Improvements are suggested to the base algorithm including aligning heart sounds, and the implementation of the Short-Time Fourier Transform to study the time evolution of the energy in the signal.

RESULTS

The improvements result in better heart beat alignment and better detection and measurement of the random events in the heart signals, so that they may provide a method to quantify nondeterministic events in heart signals. The use of the Short-Time Fourier Transform allows the examination of the random events in both time and frequency allowing for further investigation and interpretation of the signal.

CONCLUSIONS

The presented algorithm does allow for the quantification of nondeterministic events but proper care in signal acquisition and processing must be taken to obtain meaningful results.