Source Separation of the Second Heart Sound Using Gaussian Mixture Models

This paper presents an automatic normal and abnormal heart sound classification model developed based on deep learning algorithm. MITHSDB heart sounds datasets obtained from the 2016 PhysioNet/Computing in Cardiology Challenge database were used in this research with the assumption that the electrocardiograms (ECG) were recorded simultaneously with the heart sounds (phonocardiogram, PCG). The PCG time series are segmented per heart beat, and each sub-segment is converted to form a square intensity matrix, and classified using convolutional neural network (CNN) models. This approach removes the need to provide classification features for the supervised machine learning algorithm. Instead, the features are determined automatically through training, from the time series provided. The result proves that the prediction model is able to provide reasonable and comparable classification accuracy despite simple implementation. This approach can be used for real-time classification of heart sounds in Internet of Medical Things (IoMT), e.g. remote monitoring applications of PCG signal.

Acoustic stethoscopes operate on the transmission of sound from the chest piece, via air-filled hollow tubes, to the listener's ears. In the stethoscope, the chest piece and the connecting cable are known to facilitate transmission of pathogens from patient to patient and from patient to the user. Replacing the connecting cable with a wireless system may help reduce the potential risk and further allow broadcasting of the signals to multi-user for examination. So, we present a wireless stethoscope design with mobile integration that transmits heart sounds to mobile devices for evaluation and analysis, thus, eliminates the connecting tube. In this project, the chest-piece of the traditional stethoscope is integrated with micro-controller unit and Bluetooth communication device. Captured signals are processed and transmitted wireless to a mobile device with interface application software for recording, listening and visual display of wave forms . This system may serve as the means for monitoring patients from remote locations and can also be useful for training health personnel through broadcasting of recorded signals for analysis and evaluation by members of a medical team. Keywords: Digital wireless stethoscope, Cardiac auscultation, Mobile integration, Micro-controller unit, Bluetooth communication device. Cite this Article Sampada Waman, Siddhesh Sawant, Pranita Mhatre, Rutesh Tele, Prof. Sarita Kale. Bluetooth Based Wireless Digital Stethoscope with Mobile Integration. Trends in Opto-electro & Optical Communication . 2020; 10(2): 1–4p.

Regional ventilation and changes in airway caliber have been studied by analysis of breath sounds. The aim of this study was to find which frequency band is most suitable for such purposes. The subjects were 19 healthy men. Breath sounds at 6 sites on the chest wall, airflow rate at the mouth, and ECG were digitized and recorded for 30 seconds. Power spectra of breath sounds at various frequency bands (10-50 Hz, 50-100 Hz, 100-150 Hz, ...., 950-1000 Hz) were calculated with a fast-Fourier transform (FFT) for every block of 512 points (102.4 msec), and mean airflow rates for the blocks were calculated. Data recorded during inspiration when airflow ranged between 0.5 L/s and 3 L/s were analyzed. For frequency bands from 150 Hz to 850 Hz the logarithms of the power spectra were linearly correlated with the logarithms of the airflow rates, and the correlation coefficients exceeded 0.8, but for lower frequency bands the correlations were poor. When the data contaminated by heart sounds and those recorded late in inspiration were excluded the correlations at frequency bands below 150 Hz improved. The slope of log (power)/log (flow) was about 4 for the bands from 100 to 300 Hz, but it became steeper (4 to 6) for higher frequency bands. This means that the power spectra were proportional to the fourth power of the airflow rate for bands below 300 Hz, but the relation was from the fourth to the sixth power for higher frequency bands.(ABSTRACT TRUNCATED AT 250 WORDS)

An early diastolic flow from the left ventricular apex to the base can be shown in patients with hypertrophic cardiomyopathy (HCM). This tiny flow or a diastolic paradoxic jet flow is important to detect on echocardiography because of its association with cardiovascular adverse events. We report an asymptomatic 44-year-old man with mid-ventricular obstructive HCM, in which a diastolic paradoxic jet flow was observed not only in the left ventricle but also in the right ventricle. The diastolic paradoxic jet flow in the right ventricle started approximately 110 ms after the onset of the second heart sound, lasted for almost 95 ms, and disappeared in coincidence with the third heart sound; the onset was later and the duration was similar, compared with the diastolic paradoxic jet flow in the left ventricle. He had been doing well without any medication for months and later lost to follow-up. J Med Cases. 2020;11(3):57-60 doi: https://doi.org/10.14740/jmc3210

In this paper, we propose a novel biometric method based on the Heart sounds and the Gaussian Mixture Model (GMM). Heart sounds are trained by GMM to build an identification system. The MFCC Feature extraction algorithm is studied and GMM model is built. The optimal parameters are achieved by varying experimental parameters. The system has an accurate recognition rate up to 100% under the experimental conditions. The results show that the system based on GMM has a better performance than the system based on Vector Quantization (VQ).

It's to introduce the structure and operating principle of two electronic stethoscopes,one with a simple noise-stop pickup,while the other one with an active inverting denoiser.By subjective auscultation comparison and biological sounds processing via spectral view software,it got visual effects and made objective comparison.The new active inverting denoised stethoscope,the circuit of which is stable and reliable,can get a better denoised effect than the simple noise-stop pickup stethoscope.The new one is easy to operate,and can display the heart sounds,breath sounds and bowel sounds.It can solve the problem of external noise interfering auscultation better than other means. Key words: Inverting denoised; Electronic stethoscope; Spectral view

This paper is concerned with the analysis of the first (S1) and second (S2) heart sound of the Phonocardiogram signal (PCG) using Discrete Wavelet Transform (DWT) and Continuous Wavelet Transform (CWT). The second heart sound S2 consists of two major components A2 and P2. The time delay between them plays very vital role in medical diagnosis. DWT is used for denoising and finding the best split of A2 and P2 components of S2, where as, CWT is used to determine the number of frequency components of S1 and S2. Experiments are performed on normal and pathological PCG signals. Denoising and finding split between A2 and P2 is carried out using DWT. The frequency components of S1 and S2 of PCG are determined using CWT. Also split between A2 and P2 have been measured using CWT.

In this work, gender detection was carried out using heart sounds of each persons. The proposed method has three stages to make detection gender with heart sound. First, some features are obtained from heart sounds. Second, modelling is made by Gauss Mixture Model (GMM) by using obtained these features and models are trained. Finally, tested person is decided to be he or she with likelihood ratio test.

High quality speech-to-lips conversion, investigated in t his work, renders realistic lips movement (video) consistent with input sp eech (audio) without knowing its linguistic content. Instead of memoryless framebased conversion, we adopt maximum likelihood estimation of the visual parameter trajectories using an audio-visual joint Gau ssian Mixture Model (GMM). We propose a minimum converted trajectory error approach (MCTE) to further refine the converted visual paramete rs. First, we reduce the conversion error by training the joint audio-visual GMM with weighted audio and visual likelihood. Then MCTE uses the generalized probabilistic descent algorithm to minimize a conversion error of the visual parameter trajectories defined on the optimal Gau ssian kernel sequence according to the input speech. We demonstrate the effectiveness of the proposed methods using the LIPS 2009 Visual Speech Synthesis Challenge dataset, without knowing the linguistic (phonetic) content of the input speech. Index Terms: visual speech synthesis, speech-to-lips conversion, mini mum conversion error, minimum generation error

Although auscultation of heart sounds is a commonly used diagnostic tool in preliminary investigation of heart function, no description of normal amplitudes of the two major heart sounds, the first (S1) and second (S2) heart sounds, and their relation to age, sex, and body mass index (BMI), exist. Heart sound recordings from 739 healthy subjects, from the Dan-NICAD study, recorded with Acarix CADScor System at the left fourth inter-costal space (IC4), were investigated. The S1 and S2 sounds were isolated using an automated segmentation algorithm and their absolute amplitudes, |S1| and |S2|, were tested for correlation with age, sex, and BMI. Mean S11 was 95.4 (±6.2) dB for men and 93.7 (±6.8) dB for women, which showed a significant difference (p<0.05). Mean |S2| was 91.9 (±5.1) dB for men and 91.8 (±5.6) dB for women, showing no significant difference (p=0.82). |S1| and |S2| were only weakly correlated with age (r=−0.084 and r=−0.069 respectively). Both |S1| and |S2| were well correlated with BMI (r=−0.42 and r=−0.50 respectively). The correlation between BMI and the amplitude of heart sounds is an important baseline knowledge when developing algorithms for automated analysis of heart sounds.

A novel methodology for evaluation of [formula omitted] wide split via estimated parameters

We develop a general framework for compressive linear-projection measurements with side information. Side information is an additional signal correlated with the signal of interest. We investigate the impact of side information on classification and signal recovery from low-dimensional measurements. Motivated by real applications, two special cases of the general model are studied. In the first, a joint Gaussian mixture model is manifested on the signal and side information. The second example again employs a Gaussian mixture model for the signal, with side information drawn from a mixture in the exponential family. Theoretical results on recovery and classification accuracy are derived. The presence of side information is shown to yield improved performance, both theoretically and experimentally.

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.

Biometric recognition refers to the automatic recognition of individuals based on their physiological or behavioral characteristics. This paper presents a new biometric identification technique based on heart sounds. It is a signal of benefits recognition by analyzing the basic characteristics of heart sound. For first heart sound signal s1, secondly heart sound signal s2, their characteristic parameters are extracted by the independent sub-function and wavelet transform. And the heart sound cycle-power-frequency drawing (CPFD) is used to achieve identity recognition of single channel heart sound signal. The data of multi-feature extraction was processed based on the data layer fusion and improved D-S information fusion method. The results of practical experiments show the effectiveness of the heart sound identification technology which has a promising future.

Introduction: Dilated cardiomyopathy (DCM) is a progressive disease of the heart muscle that is characterized by ventricular chamber enlargement and contractile dysfunction. This condition is the third most common cause of heart failure and the most frequent reason for heart transplantation. Materials and methods: This is a clinical case of a 61 year old male diagnosed with DCM a few years ago with manifestations of heart failure. The patient is admitted with symptoms of easy fatigue, pro­gressive dyspnea, cough and paroxysmal nocturnal dyspnea-orthopnea. He used to be a professional athlete. Present symptoms came up after a virus infection with fever up to 40°C. Results: On the physical examination he was with regular rhythm, frequent extrasystoles, tachycar­dia, S3 gallop. Echocardiogram showed dilated heart cavities, reduced ejection fraction of 34%. Ther­apy with diuretics, anticoagulants, beta-blockers, angiotensin II receptor blockers was started. Fif­teen days after dehospitalization the symptoms persisted. He was discussed for cardiac resynchro­nization therapy (CRT), but the QRS complex wasn`t widened enough. He met the criteria for ther­apy with a new drug class - angiotensin receptor neprilysin inhibitor. It combines the angiotensin II receptor blocker Valsartan and the neprilysin inhibitor prodrug Sacubitril in a 1:1 ratio. Sacubi­tril is converted by esterase, which inhibits neprilysin, the enzyme responsible for the degradation of the natriuretic peptides and many other vasoactive peptides. This combination addresses two of the pathophysiological mechanisms of heart failure: activation of the renin-angiotensin-aldosterone sys­tem and decreased sensitivity to natriuretic peptides. Since then his general condition improved and the symptoms of the heart failure decreased. Conclusions: DCM is very common and irreversible disease. The clinical case is also an example of the good effect of this new class of drugs on patients with heart failure.