Heart Sound Components Research Articles

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 automated tool for localization of heart sound components S1, S2, S3 and S4 in pulmonary sounds using Hilbert transform and Heron’s formula

The primary problem with lung sound (LS) analysis is the interference of heart sound (HS) which tends to mask important LS features. The effect of heart sound is more at medium and high flow rate than that of low flow rate. Moreover, pathological HS obscures LS in a higher degree than normal HS. To get over this problem, several HS reduction techniques have been developed. An important preprocessing step in HS reduction is localization of HS components. In this paper, a new HS localization algorithm is proposed which is based on Hilbert transform (HT) and Heron’s formula. In the proposed method, the HS included segment is differentiated from the HS excluded segment by comparing their area with an adaptive threshold. The area of a HS component is calculated from the Hilbert envelope using Heron’s triangular formula. The method is tested on real recorded and simulated HS corrupted LS signals. All the experiments are conducted under low, medium and high breathing flow rates. The proposed method shows a better performance than the comparative Singular Spectrum Analysis (SSA) based method in terms of accuracy (ACC), detection error rate (DER), false negative rate (FNR), and execution time (ET).

Quasi-periodic modeling for heart sound localization and suppression in lung sounds

We present a novel approach to treating lung sounds contaminated by heart sounds by means of quasi-periodic signal modeling. Heart sounds are described through the long and short-term quasi-periodicity generated by cardiac cycle repetition and oscillations caused by valves openings and closures, respectively. In terms of signals, these quasi-periodicities drive time-variant HS envelope and phase which are modeled by single and piecewise time polynomials, respectively. Single polynomials account for slow and continuous envelope time variations, while piecewise polynomials capture fast and abrupt phase changes in short time intervals. Such a compact signal description provides an efficient way to fundamental heart sound (FHS) components localization and posterior removal from lung sounds. The results show that the proposed method outperforms two reference methods for medium (15ml/s/kg) and high (22.5ml/s/kg) air flow rates.

Measurement of two new indicators of cardiac reserve in humans, rats, rabbits, and dogs

Background: It is difficult to observe the phenomena of cardiac fatigue under a low or moderate workload test, and little was reported about cardiac fatigue under a heavy workload test because of the potential risk (sudden death) of prolonged strenuous exercise. Animal experiments would be helpful to resolve this problem. The objective of this study was to obtain baseline data of two new indicators of cardiac reserve in rats, rabbits, and dogs. Methods: Ten New Zealand White rabbits (1. 5 to 2 kg, 4 females), 10 male Wistar rats (2 months old), and 10 Golden Retriever dogs (40 days old, 5 females) were included in this study. Phonocardiogram of each animal was recorded. The basic points concerning heart sound quantitative analysis were: 1) measuring the duration and the amplitude of relevant heart sound components; 2) calculating and analyzing relevant indicators based on the data obtained from the above measurements, including the ratio of diastolic to systolic duration (D/S) and the ratio of the amplitude of the first heart sound to the amplitude of the second heart sound (S1/S2). Results: The baseline data of D/S ratio and S1/S2 ratio in rats, rabbits, and dogs were obtained. The swimming time to exhaustion for rabbits was several to a dozen minutes, and for rats, several hours. Conclusion: D/S ratio has an important biological implication, which is a safe and easy indicator for evaluating the cardiac health status of both animals and humans.

Detection of the third and fourth heart sounds using Hilbert-Huang transform

BackgroundThe third and fourth heart sound (S3 and S4) are two abnormal heart sound components which are proved to be indicators of heart failure during diastolic period. The combination of using diastolic heart sounds with the standard ECG as a measurement of ventricular dysfunction may improve the noninvasive diagnosis and early detection of myocardial ischemia.MethodsIn this paper, an adaptive method based on time-frequency analysis is proposed to detect the presence of S3 and S4. Heart sound signals during diastolic periods were analyzed with Hilbert-Huang Transform (HHT). A discrete plot of maximal instantaneous frequency and its amplitude was generated and clustered. S3 and S4 were recognized by the clustered points, and performance of the method was further enhanced by period definition and iteration tracking.ResultsUsing the proposed method, S3 and S4 could be detected adaptively in a same method. 90.3% of heart sound cycles with S3 were detected using our method, 9.6% were missed, and 9.6% were false positive. 94% of S4 were detected using our method, 5.5% were missed, and 16% were false positive.ConclusionsThe proposed method is adaptive for detecting low-amplitude and low-frequency S3 and S4 simultaneously compared with previous detection methods, which would be practical in primary care.

Beat-to-beat systolic time-interval measurement from heart sounds and ECG

Systolic time intervals are highly correlated to fundamental cardiac functions. Several studies have shown that these measurements have significant diagnostic and prognostic value in heart failure condition and are adequate for long-term patient follow-up and disease management. In this paper, we investigate the feasibility of using heart sound (HS) to accurately measure the opening and closing moments of the aortic heart valve. These moments are crucial to define the main systolic timings of the heart cycle, i.e. pre-ejection period (PEP) and left ventricular ejection time (LVET). We introduce an algorithm for automatic extraction of PEP and LVET using HS and electrocardiogram. PEP is estimated with a Bayesian approach using the signal's instantaneous amplitude and patient-specific time intervals between atrio-ventricular valve closure and aortic valve opening. As for LVET, since the aortic valve closure corresponds to the start of the S2 HS component, we base LVET estimation on the detection of the S2 onset. A comparative assessment of the main systolic time intervals is performed using synchronous signal acquisitions of the current gold standard in cardiac time-interval measurement, i.e. echocardiography, and HS. The algorithms were evaluated on a healthy population, as well as on a group of subjects with different cardiovascular diseases (CVD). In the healthy group, from a set of 942 heartbeats, the proposed algorithm achieved 7.66 ± 5.92 ms absolute PEP estimation error. For LVET, the absolute estimation error was 11.39 ± 8.98 ms. For the CVD population, 404 beats were used, leading to 11.86 ± 8.30 and 17.51 ± 17.21 ms absolute PEP and LVET errors, respectively. The results achieved in this study suggest that HS can be used to accurately estimate LVET and PEP.

Localizing Heart Sounds in Respiratory Signals Using Singular Spectrum Analysis

Respiratory sounds are always contaminated by heart sound interference. An essential preprocessing step in some of the heart sound cancellation methods is localizing primary heart sound components. Singular spectrum analysis (SSA), a powerful time series analysis technique, is used in this paper. Despite the frequency overlap of the heart and lung sound components, two different trends in the eigenvalue spectra are recognizable, which leads to find a subspace that contains more information about the underlying heart sound. Artificially mixed and real respiratory signals are used for evaluating the performance of the method. Selecting the appropriate length for the SSA window results in good decomposition quality and low computational cost for the algorithm. The results of the proposed method are compared with those of well-established methods, which use the wavelet transform and entropy of the signal to detect the heart sound components. The proposed method outperforms the wavelet-based method in terms of false detection and also correlation with the underlying heart sounds. Performance of the proposed method is slightly better than that of the entropy-based method. Moreover, the execution time of the former is significantly lower than that of the latter.

Abstract 4865: The Amplitude Ratio of the First to Second Heart Sound Mirrors Left Ventricular Systolic Function

Background: In animal studies, the amplitude of the mitral component of the first heart sound (S1) is proportional to the rate of left ventricular pressure rise (dP/dt). To develop a clinical application for this property, we tested the hypothesis that the relative amplitudes of S1 and S2 (aortic component of second heart sound) represent left ventricular dP/dt and ejection fraction (EF) in humans. Methods: Phono-electrocardiographic recordings using a digital hand-held device (HD Medical Group Ltd) followed by an echocardiogram were performed within two hours of a clinically indicated cardiac catheterization. Left ventricular dP/dt was estimated by intracardiac manometer or Doppler echocardiogram, left ventricular EF was measured by biplane 2D. The largest positive to negative deflections of the first and second heart sound components were defined as S1 and S2. The ratios of S1 to S2 (S1/S2) were obtained from the right upper sternal border (RUSB), left upper sternal border (LUSB) and apical area. Normal EF and dP/dt were defined as ≥ 55% and ≥ 1000 mmHg/s respectively. Results: We studied 61 patients with dP/dt ranging from 400 to 2850 mmHg/sec, EF ranging from 11 to 78 %. Body mass index, P-R interval, pulmonary or systemic blood pressure were not different between the normal and reduced EF or dP/dt groups. Compared with the group with reduced dP/dt or EF, the median S1/S2 at LUSB and RUSB were significantly higher in those with normal dP/dt (1.52 vs 0.78 for LUSB; 1.31 vs 0.75 for RUSB, p<0.05) or EF (1.44 vs 0.95 for LUSB; 1.65 vs 0.83 for RUBS, p<0.05). On receiver operating curve analysis, S1/S2 at LUSB and RUSB, had significant discriminative values for detecting normal dP/dt (AUC = 0.82 for LUSB, 0.74 for RUSB) and EF (AUC = 0.72 for LUSB, 0.68 for RUSB). The sensitivity and specificity of S1/S2 ≥1 for detecting normal left ventricular systolic function (normal dP/dt or EF) were 72– 80% and 57–71% respectively. Conclusion: This study demonstrated that S1, corrected for S2, mirrors left ventricular systolic function. A ratio of S1 to S2 of ≥1 is a useful indicator of normal left ventricular systolic function defined by dP/dt ≥1000 mmHg/s or EF ≥55%. Digital phonocardiography appears promising as an adjunctive bed-side tool for assessing left ventricular systolic function.

Wavelet-based Segmentation and Feature Extraction of Heart Sounds for Intelligent PDA-based Phonocardiography

Many pathological conditions of the cardiovascular system cause murmurs and aberrations in heart sounds. Phonocardiography provides the clinician with a complementary tool to record the heart sounds heard during auscultation. The advancement of intracardiac phonocardiography combined with modern digital signal processing techniques has strongly renewed researchers' interest in studying heart sounds and murmurs. The aim of this work is to investigate the applicability of different spectral analysis methods to heart sound signals and explore their suitability for PDA-based implementation. Fourier transform (FT), short-time Fourier transform (STFT) and wavelet transform (WT) are used to perform spectral analysis on heart sounds. A segmentation algorithm based on Shannon energy is used to differentiate between first and second heart sounds. Then wavelet transform is deployed again to extract 64 features of heart sounds. The FT provides valuable frequency information but the timing information is lost during the transformation process. The STFT or spectrogram provides valuable time-frequency information but there is a trade-off between time and frequency resolution. Wavelet analysis, however, does not suffer from limitations of the STFT and provides adequate time and frequency resolution to accurately characterize the normal and pathological heart sounds. The results show that the wavelet-based segmentation algorithm is quite effective in localizing the important components of both normal and abnormal heart sounds. They also demonstrate that wavelet-based feature extraction provides suitable feature vectors which are clearly differentiable and useful for automatic classification of heart sounds.

Heart-sound detecting apparatus and heart-sound detecting method

An apparatus for detecting a heart sound of a living subject, including a pressure-pulse-wave sensor which is adapted to be worn on a body portion of the subject that is distant from a chest of the subject, detects a pressure pulse wave produced by an artery of the body portion, and generates a pressure-pulse-wave signal representing the detected pressure pulse wave; and a heart-sound extracting device for extracting, from the pressure-pulse-wave signal generated by the pressure-pulse-wave sensor, a heart-sound component representing the heart sound of the subject.

Heart-sound detecting apparatus

An apparatus for detecting a heart sound of a living subject, including a memory device which stores heart-sound characteristic information which is characteristic of a heart sound of the subject, a heart-sound sensor which is adapted to be worn on a body portion of the subject that is distant from a chest of the subject and which detects, from the body portion, a physical signal containing a heart-sound component and supplies the physical signal, and a heart-sound determining device for determining, based on the heart-sound characteristic information stored in the memory device, the heart-sound component contained in the physical signal.

Classification of timing and loudness of second heart sound components by expert and novice physicians

Consistent and accurate assessment of the relative timing and loudness of second heart sound (S2) components is essential to skilled performance of cardiac auscultation. Classification of S2 “splitting” and loudness of P2 (the pulmonic component of S2) by expert and novice physicians were examined in experiments designed to minimize the influence of higher order cognitive factors on auscultation performance. Stimuli were recorded heart sounds in which the temporal separation and relative intensity of S2 components were controlled by digital wave form editing. The task was to classify S2 splitting and P2 loudness using categories commonly employed in making chart notation. Experts' responses were more consistent and accurate with respect to detection of underlying pathology than the responses of novices. Perceptual category boundaries for novices varied when the heart rate changed and when a systolic murmur was introduced, while expert boundaries remained stable. Most experts were able to detect asynchrony in S2 components of less than 20 ms. [Work supported by the Graduate School and the Dwan Chair for Education in Pediatric Cardiology at the University of Minnesota.]

A phonocardiographic study of equine heart sounds.

The occurrence and timing of heart sounds were examined from phonocardiograms taken from the mitral, aortic and tricupsid recording areas in each of 18 horses. 10 sound events could be identified with each cardiac cycle. Atrial contraction produced up to 3 sound events. The first heart sound consisted of 4 components whereas the second sound was single. Two sound events were associated with the 3rd heart sound in early diastole. The occurrence of third and fourth heart sound components varied between horses and between recording areas. The mitral recording area was considered most satisfactory for routine phonocardiographic studies in horses.

Intensity of the normal second heart sound components in their traditional auscultatory areas

Incomplete and conflicting information on the intensity characteristics of the normal second heart sound (S2) led to a phonocardiographic study in 160 subjects ranging from six to seventy-nine years of age. Routine auscultation was simulated in 100 subjects by end-expiratory recordings with a single microphone applied parasternally first to the second right intercostal space (“aortic area”) and then to the second left intercostal space (“pulmonic area”). In the remaining sixty subjects four consecutive sets of simultaneous tracings were obtained from the two areas using a pair of calibrated microphone amplifiers. Statistical analysis of reproducibility revealed the calibrated microphone technic to be distinctly superior to the single microphone method. In 125 subjects splitting of S2 was demonstrable in the pulmonic area. In 70 per cent of these the maximal deflection of the aortic component (A2) was greater than that of the pulmonic component (P2); in 10 per cent A2 and P2 were of equal intensity; in 19 per cent there was a varying relationship in intensity between A2 and P2; and in only one subject, a child, was P2 consistently greater than A2. The maximal wave deflection of S2 (single or split) was used to compare the intensity of S2 in the two recording areas. The results suggest that although in children and young adults S2 will usually be heard better in the pulmonic area, in most adults it is equally likely that S2 will be heard better in the aortic area, or in the pulmonic area, or that S2 will be of equal intensity in the two areas. This degree of variability among different subjects and the inherent variability of the routine method of auscultation itself, especially in comparing sound intensity in two areas, would tend to negate any clinical usefulness of this comparison.

Normal first heart sounds with nonfunctional tricuspid valve or right ventricle. Clinical and experimental evidence.

A case of Ebstein's malformation of the tricuspid valve with persistent foramen secundum and right bundle-branch block is presented. Preoperative tracings revealed three distinct components of the first sound with normal intervals between them. After resection of a single floating tricuspid leaflet, closure of the septal defect, and insertion of a Starr-Edwards valve, the phonocardiogram again showed three normal components of the first heart sound, followed by a loud closing snap of the prosthetic tricuspid valve. This case demonstrates the occurrence of the three normal components of the first heart sound independently of right ventricular contraction and of tricuspid closure or tensing. In five dogs left ventricular phonocardiograms, made after right heart bypass and destruction of the free ventricular wall, revealed three normal components of the first heart sounds in three animals, and of the first two components, separated by normal intervals in two others. These experiments exclude any participation of the right ventricle or tricuspid valve in the mechanism of origin of the first heart sound. The authors conclude that the three normal components of the first heart sound are of left-sided origin and that the designation "tricuspid" for the second component should be abandoned.

Studies on the Intracardiac Phonocardiography

The intracardiac phonocardiographic method of our original conception has been an epock making advance in the study of heart sounds. In this paper with the accumulation of 50 additional clinical cases from the time of original reports the result are summarized and some extra views are expressed. 1) Our equipment utilizing the condensor-type pick up is believed to be for superior to those made and used in Europe and the United States from the view-point of sensitivity and frequency response when compared by actual use. 2) The characteristics of intracardiac sounds were discussed by grouping the deta under the headings of “the effect of blood flow”, “damping due to heart and blood-vessel wall”, “conduction of vibrations”, “Knocking sound”, etc. 3) The various components of the normal heart sounds were divided into the first sound, 2nd sound, 3rd sound, atrial sound, extra sound and murmurs. The results of the clinical and animal experiments were discussed. 4) The advantages of this method for determining the region of the murmur origin, its mechanisms and in clarifying the local blood movement within the heart and its importance as a diagnostic tool were discussed.