Hemodynamics-driven mathematical model of first and second heart sound generation.

NEW & NOTEWORTHY: To the best of our knowledge, this is the first hemodynamic-based heart sound generation model embedded in a complete real-time computational model of the cardiovascular system. Simulated heart sounds are similar to experimental and clinical measurements, both quantitatively and qualitatively. Our model can be used to investigate the relationships between heart sound acoustic features and hemodynamic factors/anatomical parameters.

Classification of heart sounds based on the combination of the modified frequency wavelet transform and convolutional neural network.

We purpose a novel method that combines modified frequency slice wavelet transform (MFSWT) and convolutional neural network (CNN) for classifying normal and abnormal heart sounds. A hidden Markov model is used to find the position of each cardiac cycle in the heart sound signal and determine the exact position of the four periods of S1, S2, systole, and diastole. Then the one-dimensional cardiac cycle signal was converted into a two-dimensional time-frequency picture using the MFSWT. Finally, two CNN models are trained using the aforementioned pictures. We combine two CNN models using sample entropy (SampEn) to determine which model is used to classify the heart sound signal. We evaluated our model on the heart sound public dataset provided by the PhysioNet Computing in Cardiology Challenge 2016. Experimental classification performance from a 10-fold cross-validation indicated that sensitivity (Se), specificity (Sp) and mean accuracy (MAcc) were 0.95, 0.93, and 0.94, respectively. The results showed the proposed method can classify normal and abnormal heart sounds with efficiency and high accuracy. Graphical abstract Block diagram of heart sound classification.

Analysis of PCG signals using quality assessment and homomorphic filters for localization and classification of heart sounds.

BACKGROUND AND OBJECTIVE: Accurate localization of heart beats in phonocardiogram (PCG) signal is very crucial for correct segmentation and classification of heart sounds into S1 and S2. This task becomes challenging due to inclusion of noise in acquisition process owing to number of different factors. In this paper we propose a system for heart sound localization and classification into S1 and S2. The proposed system introduces the concept of quality assessment before localization, feature extraction and classification of heart sounds. METHODS: The signal quality is assessed by predefined criteria based upon number of peaks and zero crossing of PCG signal. Once quality assessment is performed, then heart beats within PCG signal are localized, which is done by envelope extraction using homomorphic envelogram and finding prominent peaks. In order to classify localized peaks into S1 and S2, temporal and time-frequency based statistical features have been used. Support Vector Machine using radial basis function kernel is used for classification of heart beats into S1 and S2 based upon extracted features. The performance of the proposed system is evaluated using Accuracy, Sensitivity, Specificity, F-measure and Total Error. The dataset provided by PASCAL classifying heart sound challenge is used for testing. RESULTS: Performance of system is significantly improved by quality assessment. Results shows that proposed Localization algorithm achieves accuracy up to 97% and generates smallest total average error among top 3 challenge participants. The classification algorithm achieves accuracy up to 91%. CONCLUSION: The system provides firm foundation for the detection of normal and abnormal heart sounds for cardiovascular disease detection.

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), ß, 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.

Classification of prolapsed mitral valve versus healthy heart from phonocardiograms by multifractal analysis.

Phonocardiography has shown a great potential for developing low-cost computer-aided diagnosis systems for cardiovascular monitoring. So far, most of the work reported regarding cardiosignal analysis using multifractals is oriented towards heartbeat dynamics. This paper represents a step towards automatic detection of one of the most common pathological syndromes, so-called mitral valve prolapse (MVP), using phonocardiograms and multifractal analysis. Subtle features characteristic for MVP in phonocardiograms may be difficult to detect. The approach for revealing such features should be locally based rather than globally based. Nevertheless, if their appearances are specific and frequent, they can affect a multifractal spectrum. This has been the case in our experiment with the click syndrome. Totally, 117 pediatric phonocardiographic recordings (PCGs), 8 seconds long each, obtained from 117 patients were used for PMV automatic detection. We propose a two-step algorithm to distinguish PCGs that belong to children with healthy hearts and children with prolapsed mitral valves (PMVs). Obtained results show high accuracy of the method. We achieved 96.91% accuracy on the dataset (97 recordings). Additionally, 90% accuracy is achieved for the evaluation dataset (20 recordings). Content of the datasets is confirmed by the echocardiographic screening.

Time-frequency analysis of phonocardiogram signals using wavelet transform: a comparative study.

Analysis of phonocardiogram (PCG) signals provides a non-invasive means to determine the abnormalities caused by cardiovascular system pathology. In general, time-frequency representation (TFR) methods are used to study the PCG signal because it is one of the non-stationary bio-signals. The continuous wavelet transform (CWT) is especially suitable for the analysis of non-stationary signals and to obtain the TFR, due to its high resolution, both in time and in frequency and has recently become a favourite tool. It decomposes a signal in terms of elementary contributions called wavelets, which are shifted and dilated copies of a fixed mother wavelet function, and yields a joint TFR. Although the basic characteristics of the wavelets are similar, each type of the wavelets produces a different TFR. In this study, eight real types of the most known wavelets are examined on typical PCG signals indicating heart abnormalities in order to determine the best wavelet to obtain a reliable TFR. For this purpose, the wavelet energy and frequency spectrum estimations based on the CWT and the spectra of the chosen wavelets were compared with the energy distribution and the autoregressive frequency spectra in order to determine the most suitable wavelet. The results show that Morlet wavelet is the most reliable wavelet for the time-frequency analysis of PCG signals.

Noninvasive prediction of left ventricular systolic dysfunction in patients with clinically suspected heart failure using acoustic cardiography.

The signs and symptoms of systolic heart failure are frequently insensitive and nonspecific, making an accurate bedside diagnosis of left ventricular systolic dysfunction (LVSD) challenging. B-type natriuretic peptide (BNP) is often used, but is not diagnostically useful when in the indeterminate range. The authors investigated the diagnostic test characteristics of acoustic cardiographic parameters to identify patients with LVSD. Four hundred thirty-three patients with contemporaneous measurements of computerized acoustic cardiography, BNP, and echocardiography were included. The acoustic cardiographic model outperformed BNP alone at detecting reduced left ventricular ejection fraction (C statistic, 0.88 vs 0.67; P<.0001). The acoustic model with BNP did not perform better than the acoustic model alone (P=.14). Within the indeterminate BNP range, the acoustic model outperformed BNP (C statistic, 0.89 vs 0.64; P<.0001). Noninvasive computerized acoustic cardiography predicted LVSD in a diverse population. This acoustic cardiographic model outperformed BNP alone for predicting LVSD.

Autonomous auscultation of the human heart employing a precordial electro-phonocardiogram and ensemble empirical mode decomposition.

The research presented in this paper serves to provide a tool to autonomously screen for cardiovascular disease in the rural areas of Africa. With this tool, cardiovascular disease can potentially be detected in its initial stages, which is essential for effective treatment. The autonomous auscultation system proposed here utilizes recorded heart sounds and electrocardiogram signals to automatically distinguish between normal and abnormal heart conditions. Patients that are identified as abnormal by the system can then be referred to a specialist consultant, which will save a lot of unnecessary referrals. In this study, heart sound and electrocardiogram signals were recorded with the prototype precordial electro-phonocardiogram device, as part of a clinical study to screen patients for cardiovascular disease. These volunteers consisted of 28 patients with a diagnosed cardiovascular disease and, for control purposes, 34 persons diagnosed with healthy hearts. The proposed system employs wavelets to first denoise the recorded signals, which is then followed by segmentation of heart sounds. Frequency spectrum information was extracted as diagnostic features from the heart sounds by means of ensemble empirical mode decomposition and auto regressive modelling. The respective features were then classified with an ensemble artificial neural network. The performance of the autonomous auscultation system used in concert with the precordial electro-phonocardiogram prototype showed a sensitivity of 82% and a specificity of 88%. These results demonstrate the potential benefit of the precordial electro-phonocardiogram device and the developed autonomous auscultation software as a screening tool in a rural healthcare environment where large numbers of patients are often cared for by a small number of inexperienced medical personnel.

The moment segmentation analysis of heart sound pattern.

UNLABELLED: This paper presents two new ideas. The first one is to apply the Viola integral waveform method to analyze the heart sounds recorded by an electric stethoscope, and the multi-scale moment analysis is proposed to locate each cycle of heart sounds. A fast algorithm for calculating characteristic waveform (CW) and characteristic moment waveform (CMW) of heart sound can be expressed by the Viola integral method, and their calculation time has nothing to do with their scales. The second idea is easier to segment the heart sound based on its approximate cyclical characteristic than the ordinary methods. Each heart sound cycle can be quickly found by CMW's Local Extreme Points (LEPs). Based on the information of LEPs and CW, a high accurate search algorithm to segment S1 and S2 sounds is submitted. By numerical experiments, the important parameters of time scale delta=0.05s for CW and l=0.45s for CMW are obtained and validated for segmentation of heart sound. CONCLUSION: More exact segmentation boundaries of the heart sound signal could be located fast in an automated way, and a further performance analysis is presented. Owing to the use of the rhythm of CMW curves, the proposed method not only gives a higher success segmentation rate, but also it is actually simpler and faster than the wavelet method.

Real-time, low-complexity, low-memory solution to ECG-based heart rate detection.

This paper addresses the issue of heart rate detection from noisy ECG data, and presents a method with low complexity and low memory requirements that can detect QRS complex in the presence of noise and muscle artifacts. On the MIT-BIH arrhythmia database we were able to detect 99.3% of QRS complexes with 0.47% false detection. This method can also be applied to heart rate detection using phonocardio signals.

Computer method for perinatal screening of cardiac murmur using fetal phonocardiography.

The main purpose of this paper is to demonstrate the capability of fetal phonocardiographic measurements to indicate some congenital heart defects. It deals with the results of investigations carried out during the last four years involving 820 pregnant women. During the investigations fetal cardiac murmurs presenting typical waveforms and incidences of acoustic signals were recorded. Causes of these murmurs are suggested based on comparison with the well-known waveforms of infants and children. A sophisticated signal processing method for murmur discovery is presented, that is also useful for automatic perinatal screening after the 28th week of gestation. By these means low-risk population may also be fully tested for cardiac malfunctions.

Feature extraction for murmur detection based on support vector regression of time-frequency representations.

This paper presents a nonlinear approach for time-frequency representations (TFR) data analysis, based on a statistical learning methodology - support vector regression (SVR), that being a nonlinear framework, matches recent findings on the underlying dynamics of cardiac mechanic activity and phonocardiographic (PCG) recordings. The proposed methodology aims to model the estimated TFRs, and extract relevant features to perform classification between normal and pathologic PCG recordings (with murmur). Modeling of TFR is done by means of SVR, and the distance between regressions is calculated through dissimilarity measures based on dot product. Finally, a k-nn classifier is used for the classification stage, obtaining a validation performance of 97.85%.

Phonocardiographic signal analysis method using a modified hidden Markov model.

Auscultation is an important diagnostic indicator for cardiovascular analysis. Heart sound classification and analysis play an important role in the auscultative diagnosis. This study uses a combination of Mel-frequency cepstral coefficient (MFCC) and hidden Markov model (HMM) to efficiently extract the features for pre-processed heart sound cycles for the purpose of classification. A system was developed for the interpretation of heart sounds acquired by phonocardiography using pattern recognition. The task of feature extraction was performed using three methods: time-domain feature, short-time Fourier transforms (STFT) and MFCC. The performances of these feature extraction methods were then compared. The results demonstrated that the proposed method using MFCC yielded improved interpretative information. Following the feature extraction, an automatic classification process was performed using HMM. Satisfactory classification results (sensitivity > or =0.952; specificity > or =0.953) were achieved for normal subjects and those with various murmur characteristics. These results were based on 1398 datasets obtained from 41 recruited subjects and downloaded from a public domain. Constituents characteristics of heart sounds were also evaluated using the proposed system. The findings herein suggest that the described system may have the potential to be used to assist doctors for a more objective diagnosis.

A dynamic method to estimate the time split between the A2 and P2 components of the S2 heart sound.

The time interval between the aortic (A2) and the pulmonary (P2) components of the second heart sound (S2) is an indicator of pulmonary arterial pressure. However, knowledge of the A2 and P2 components of the S2 sound is difficult to obtain due to their temporal overlap and significant spectral similarity. In this work, we aim to extract the A2 and P2 components from the phonocardiogram to estimate the time interval between them. We attain our objective by first isolating the S2 sound from the phonocardiogram by utilizing the mode complexity of the heart. Then, we assume the statistical independence of the A2 and P2 components and extract them from the S2 sound by the application of blind source separation techniques. Once separated, the time interval between the A2 and P2 components is estimated with a time-centroid-based method. Experimental results using simulated data show excellent performance of the proposed algorithm to extract the A2 and the P2 components from the S2 sound and to estimate the time interval between them. Results obtained from real data are also encouraging and show promise for utilizing the proposed method in a clinical setting to non-invasively tract pulmonary hypertension.

Can the timing-characteristics of phonocardiographic signal be used for cuffless systolic blood pressure estimation?

Continuous and non-invasive measurement of blood pressure (BP) is always important to critically ill patients. To achieve continuous and cuffless BP monitoring, pulse transit time (PTT) has been reported as a potential parameter. Recently a novel parameter RS2 (defined as the time interval measured from the R wave of electrocardiographic (ECG) signal to the peak of second heart sound of phonocardiographic (PCG) signal) is proposed for the same purpose. In this study, the relationship between systolic BP (SBP) and PTT as well as the relationship between SBP and RS2 on 25 healthy subjects, aged 24 +/- 3 years, were compared after exercise. The results in current study showed that SBP is correlated with both PTT and RS2, where the mean individual correlations are r=-0.95 and r=-0.85 respectively. The mean standard deviation of the differences between the measured SBP and the SBP predicted from the regression lines in scatter plots of SBP~PTT and SBP~RS2 are 4.1 mmHg and 7.2 mmHg respectively. In summary, the results showed that RS2 is possible to be used for continuous and non-invasive monitoring of SBP after exercise. In the future, it is important to investigate more robust techniques for locating characteristic points on the PCG signals.

Differences in foetal heart rate variability from phonocardiography and abdominal electrocardiography.

The agreement of the phonocardiographic method to provide foetal heart rate variability (FHRV) indices equivalent to those derived from abdominal electrocardiography was tested. 15 pregnant women were recruited in order to obtain antepartum foetal phonocardiograms and abdominal electrocardiograms three minutes long. From the respective sound (SS) and electric (RR) time series, typical temporal and spectral indices of FHRV were computed and compared. Means of the SS and RR intervals were not significant (P> 0.05) and these showed a correlation r=0.98. However, the temporal indices, P(TOT), HF and LF/HF presented differences (P< 0.05), since SS values were higher. Spectral coherence decreased below 0.5 for frequencies above 0.28+ 0.07 Hz, where bands resembling maternal and foetal breathing movements were noted. Particularly above 0.28Hz, temporal and spectral FHRV indices derived from phonocardiography and electrocardiography show differences. Quality of the signal, processing techniques, and maternal and foetal respiratory factors could account to explain these.

Beamformed nearfield imaging of a simulated coronary artery containing a stenosis.

This paper is concerned with the potential for the detection and location of an artery containing a partial blockage by exploiting the space-time properties of the shear wave field in the surrounding elastic soft tissue. As a demonstration of feasibility, an array of piezoelectric film vibration sensors is placed on the free surface of a urethane mold that contains a surgical tube. Inside the surgical tube is a nylon constriction that inhibits the water flowing through the tube. A turbulent field develops in and downstream from the blockage, creating a randomly fluctuating pressure on the inner wall of the tube. This force produces shear and compressional wave energy in the urethane. After the array is used to sample the dominant shear wave space-time energy field at low frequencies, a nearfield (i.e., focused) beamforming process then images the energy distribution in the three-dimensional solid. Experiments and numerical simulations are included to demonstrate the potential of this noninvasive procedure for the early identification of vascular blockages-the typical precursor of serious arterial disease in the human heart.

[Ventricular diastolic function: a correlation between phonomechanocardiography and echo-Doppler]. / Función diastólica ventricular: correlación entre fonomecanocardiografía y eco Doppler.

With the purpose to compare phonomechanocardiography and echo Doppler in the assessment of diastolic function of the left ventricle, we study 45 patients (30 male and 15 female) average age 50 +/- 9 years. We performed phonomechanocardiogram, echo-M, 2-D and Doppler transmitral. They were classified in four group according to mitral flow pattern: normal 14 patients; pattern I by Appleton (PI) 14 patients, 11 with aortic stenosis and 3 with hypertrophic cardiomyopathy; pattern II (PII) 12 patients with dilated cardiomyopathy grade III-IV and the last group of 5 patients with myocardial infarction with normal mitral flow but with impaired diastolic function by phonomechanocardiography. The phonomechanocardiographic index of ventricular relaxation (A2-O, ITRAT), compliance (a/D) and global diastolic function (ITAD) correlated with Doppler index (A2-D, E/A, atrial filling fraction, E-F slope and deceleration time) in N + PI group. The correlation was not significant when N + PI + PII or PI + PII groups were considered. The ITAD and E/A had r = 0.713 (p < 0.001) in N + PI, r = 0.12 (NS) in N + PI + PII and r = -0.308 (NS) in PI + PII. There was a dissociation between increased "a" wave in apexcardiogram and little "A" wave in PII patients suggesting "atrial failure". The patients with myocardial infarction received isosorbide dinitrate 5 mg showing changes of "pseudonormalizated" pattern in PI with normalized ITAD. This findings suggest that assessment of diastolic function by Doppler is dependent of loading conditions (specially preload), and cannot evaluate relaxation in PII but this is possible by phonomechanocardiography. It is advised the combination of the two technics for better assessment of diastolic function.