Effective features extraction by analyzing heart sound for identifying cardiovascular diseases related to COVID-19: A diagnostic model

Incidence and exacerbation of some of the cardiovascular diseases in the presence of the coronavirus will lead to an increase in the mortality rate among patients. Therefore, early diagnosis of such diseases is critical, especially during the COVID-19 pandemic (mild COVID-19 infection). Thus, for diagnosing the heart diseases related to the COVID-19, an automatic, non-invasive, and inexpensive method based on the heart sound processing approach is proposed. In the present study, a set of features related to the nature of heart signals is defined and extracted. The investigated features included morphological and statistical features in the heart sound frequencies. By extracting and selecting a set of effective features related to the mentioned diseases, and avoiding to use different segmentation and filtering techniques, dependence on a limited dataset and specific sampling procedures has been eliminated. Different classifiers with various kernels are applied for diagnosis in data unbalanced and balanced conditions. The results showed 93.15% accuracy and 93.72% F1-score using 60 effective features in data balanced conditions. The identification system using the extracted features from Azad dataset is able to achieve the desired results in a generalized dataset. In this way, in the shortest possible sampling time, the present system provided an effective and generalizable method and a practical model for diagnosing important cardiovascular diseases in the presence of coronavirus in the COVID-19 pandemic.

Improvement of the Accuracy in the Identification of Coronary Artery Disease Combining Heart Sound Features

Most researchers use features of diastolic murmurs to identify coronary artery disease. However, the diastolic murmurs of coronary artery disease are usually very weak and are easily contaminated by noise and valvular murmurs. Therefore, the diagnostic accuracy of coronary artery disease when only using diastolic murmurs is not well. An algorithm for improving the accuracy in the identification of coronary artery disease by combining the features of the first heart sound and diastolic murmurs was proposed. Firstly, a first heart sound feature extraction algorithm was used to identify coronary artery disease from noncoronary artery disease. Secondly, the Empirical Wavelet Transform algorithm was used to decompose the diastolic heart sound into three modes, and the spectral energy of each mode was calculated to distinguish coronary artery disease from noncoronary artery disease. Then, the features of the fist heart sound, the second diastolic spectral energy, and the parameter P3, which was used to discriminate the diastolic murmurs in coronary artery disease and in valvular disease, were combined together to improve the diagnostic accuracy of coronary artery disease. The comparison experiment results show that the accuracy of the proposed algorithm is superior to some state-of-the-art methods when they are used to diagnose coronary artery disease.

Spectral analysis of heart sounds associated with coronary artery disease

Objective. The aim of this study was to find spectral differences of diagnostic interest in heart sound recordings of patients with coronary artery disease (CAD) and healthy subjects.Approach. Heart sound recordings from three studies were pooled, and patients with clear diagnostic outcomes (positive: CAD and negative: Non-CAD) were selected for further analysis. Recordings from 1146 patients (191 CAD and 955 Non-CAD) were analyzed for spectral differences between the two groups using Welch's spectral density estimate. Frequency spectra were estimated for systole and diastole segments, and time-frequency spectra were estimated for first (S1) and second (S2) heart sound segments. An ANCOVA model with terms for diagnosis, age, gender, and body mass index was used to evaluate statistical significance of the diagnosis term for each time-frequency component.Main results. Diastole and systole segments of CAD patients showed increased energy at frequencies 20-120 Hz; furthermore, this difference was statistically significant for the diastole. CAD patients showed decreased energy for the mid-S1 and mid-S2 segments and conversely increased energy before and after the valve sounds. Both S1 and S2 segments showed regions of statistically significant difference in the time-frequency spectra.Significance. Results from analysis of the diastole support findings of increased low-frequency energy from previous studies. Time-frequency components of S1 and S2 sounds showed that these two segments likely contain heretofore untapped information for risk assessment of CAD using phonocardiography; this should be considered in future works. Further development of features that build on these findings could lead to improved acoustic detection of CAD.

Discrimination of Patients with Varying Degrees of Coronary Artery Stenosis by ECG and PCG Signals Based on Entropy

Coronary heart disease (CHD) is the leading cause of cardiovascular death. This study aimed to propose an effective method for mining cardiac mechano-electric coupling information and to evaluate its ability to distinguish patients with varying degrees of coronary artery stenosis (VDCAS). Five minutes of electrocardiogram and phonocardiogram signals was collected synchronously from 191 VDCAS patients to construct heartbeat interval (RRI)-systolic time interval (STI), RRI-diastolic time interval (DTI), HR-corrected QT interval (QTcI)-STI, QTcI-DTI, Tpeak-Tend interval (TpeI)-STI, TpeI-DTI, Tpe/QT interval (Tpe/QTI)-STI, and Tpe/QTI-DTI series. Then, the cross sample entropy (XSampEn), cross fuzzy entropy (XFuzzyEn), joint distribution entropy (JDistEn), magnitude-squared coherence function, cross power spectral density, and mutual information were applied to evaluate the coupling of the series. Subsequently, support vector machine recursive feature elimination and XGBoost were utilized for feature selection and classification, respectively. Results showed that the joint analysis of XSampEn, XFuzzyEn, and JDistEn had the best ability to distinguish patients with VDCAS. The classification accuracy of severe CHD-mild-to-moderate CHD group, severe CHD-chest pain and normal coronary angiography (CPNCA) group, and mild-to-moderate CHD-CPNCA group were 0.8043, 0.7659, and 0.7500, respectively. The study indicates that the joint analysis of XSampEn, XFuzzyEn, and JDistEn can effectively capture the cardiac mechano-electric coupling information of patients with VDCAS, which can provide valuable information for clinicians to diagnose CHD.

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.

Path length entropy analysis of diastolic heart sounds

Early detection of coronary artery disease (CAD) using the acoustic approach, a noninvasive and cost-effective method, would greatly improve the outcome of CAD patients. To detect CAD, we analyze diastolic sounds for possible CAD murmurs. We observed diastolic sounds to exhibit 1/f structure and developed a new method, path length entropy (PLE) and a scaled version (SPLE), to characterize this structure to improve CAD detection. We compare SPLE results to Hurst exponent, Sample entropy and Multiscale entropy for distinguishing between normal and CAD patients. SPLE achieved a sensitivity-specificity of 80%-81%, the best of the tested methods. However, PLE and SPLE are not sufficient to prove nonlinearity, and evaluation using surrogate data suggests that our cardiovascular sound recordings do not contain significant nonlinear properties.

Acoustic source separation for the detection of coronary artery sounds

Coronary artery disease (CAD) is the leading cause of death in the United States, being responsible for more than 20% of all deaths in the country. This is in large part due to the difficulty of diagnostic screening for CAD. Phonoangiography seeks to detect CAD via the acoustic signature associated with turbulent flow near an abnormally constricted, or stenosed, region. However, the usefulness of the technique is severely hindered by the low strength of the CAD signal compared to the background noise within the chest. In this work, acoustic finite element analysis (FEA) was performed on physiologically accurate chest geometries to demonstrate the feasibility of an original acoustic source separation methodology for isolating coronary sounds. This approach is based upon pseudoinversion of mixing matrices determined through a combination of experiment and computation. This allows calculation of the sound emitted by the coronary arteries based upon measurements of the acoustic velocity on the chest surface. This work demonstrates the feasibility of such a technique computationally and examines the vulnerability of the proposed approach to measurement errors.

Acoustic detection of coronary artery disease

Coronary artery disease (CAD) occurs when the arteries to the heart (the coronary arteries) become blocked by deposition of plaque, depriving the heart of oxygen-bearing blood. This disease is arguably the most important fatal disease in industrialized countries, causing one-third to one-half of all deaths in persons between the ages of 35 and 64 in the United States. Despite the fact that early detection of CAD allows for successful and cost-effective treatment of the disease, only 20% of CAD cases are diagnosed prior to a heart attack. The development of a definitive, noninvasive test for detection of coronary blockages is one of the holy grails of diagnostic cardiology. One promising approach to detecting coronary blockages noninvasively is based on identifying acoustic signatures generated by turbulent blood flow through partially occluded coronary arteries. In fact, no other approach to the detection of CAD promises to be as inexpensive, simple to perform, and risk free as the acoustic-based approach. Although sounds associated with partially blocked arteries are easy to identify in more superficial vessels such as the carotids, sounds from coronary arteries are very faint and surrounded by noise such as the very loud valve sounds. To detect these very weak signals requires sophisticated signal processing techniques. This review describes the work that has been done in this area since the 1980s and discusses future directions that may fulfill the promise of the acoustic approach to detecting coronary artery disease.

Spectral analysing of portal vein Doppler signals in the cirrhosis patients

In this study, we have researched the efficacy of short-time Fourier transformation (STFT) of Doppler signals from the portal veins of healthy volunteers and cirrhosis patients. Sonogram and power spectral distribution for portal vein Doppler spectral waveform changes in the cirrhosis disease were utilized and these graphics compared with healthy volunteers. Five parameters were used to compare power spectrum graphics. Clear differences were detected in the calculated parameters between healthy and cirrhosis patients. It was seen that power spectral graphics and sonograms of portal vein Doppler signals may be used to determine cirrhosis disease.

Comparison of eigenvector methods with classical and model-based methods in analysis of internal carotid arterial Doppler signals

Doppler ultrasound is known as a reliable technique, which demonstrates the flow characteristics and resistance of arteries in various vascular disease. In this study, internal carotid arterial Doppler signals recorded from 105 subjects were processed by PC-computer using classical, model-based, and eigenvector methods. The classical method (fast Fourier transform), two model-based methods (Burg autoregressive, least-squares modified Yule-Walker autoregressive moving average methods), and three eigenvector methods (Pisarenko, multiple signal classification, and Minimum-Norm methods) were selected for processing internal carotid arterial Doppler signals. Doppler power spectra of internal carotid arterial Doppler signals were obtained using these spectrum analysis techniques. The variations in the shape of the Doppler power spectra were examined in order to obtain medical information. These power spectra were then used to compare the applied methods in terms of their frequency resolution and the effects in determination of stenosis and occlusion in internal carotid arteries.

Order determination in autoregressive modeling of diastolic heart sounds

Previous studies demonstrated that spectral analysis of diastolic heart sounds may provide valuable information for the detection of coronary artery disease. Although parametric modeling methods were successfully used to achieve this goal, and showed considerable performance, the accuracy and precision of the analysis is strongly dependent on the model order selected. In order to investigate the effects of model order selection on the analysis, diastolic heart sound recorded from both normal and diseased patients were analyzed using the AR modeling, which is computationally the most efficient parametric spectral analysis method. The model order were determined by using four different model order selection criteria. The results showed that the four criteria yielded different order for the same data set. On the other hand, different criteria showed different performance in different measurement conditions. Effect of arbitrary order selection was also discussed. As a result, an optimal AR model order that may be used for every case was determined.

Investigating the effects of vasodilator drugs on the turbulent sound caused by femoral artery stenosis using short-term Fourier and wavelet transform methods

In this study, the effects of vasodilator drugs on the turbulent sound generation mechanisms during femoral artery stenoses were investigated using the wavelet analysis of the turbulent sounds to characterize these sounds before and after the injection of vasodilator drugs. Results showed that the injection of drugs drastically improved the diagnostic performance of the turbulent sounds in detecting stenoses by increasing the signal-to-noise ratio of the sounds. Results also suggested that the sound above 250 Hz was drastically increased in response to the injection of the vasodilator drug for the partially occluded cases. The turbulence sounds caused by partially occluded femoral arteries are directly related to the slope of baseline of blood flow and to the velocity of the flow. For the 0% occlusion case, initially, sounds were produced with the injection of drugs. However, the sounds totally disappeared when the slope of average blood flow was zero. These results show that the diagnostic performance of diastolic heart sounds associated with occluded arteries can be improved by using vasodilator drugs, which increase the acoustic energy in the first and second wavelet bandwidths due to the turbulence. The short-term Fourier transform (STFT) method was also applied to the same data base. Results using the STFT showed somewhat similar power distributions in that the acoustical power above 250 Hz was increased after the injection of drugs for the occluded cases. However, the WT method provided better time-frequency resolution than the STFT method, showing details of the change in the frequency characteristics with respect to time after the injection of drug.

Noninvasive characterization of the sound pattern caused by coronary artery stenosis using FTF/FAEST zero tracking filters: normal/abnormal study

In this article, a new approach has been proposed to investigate the extraction of useful information from diastolic heart sounds caused by partially occluded coronary arteries. This method, which estimates and tracks the zeros (poles) of the diastolic heart sounds directly, takes advantage of the FTF/FAEST (Fast Transversal Filters/Fast a Posteriori Error Sequential) technique which possesses the fast convergence property of the Recursive Least Square (RLS) method and the computational simplicity of the Least Mean Square (LMS) method. In previous studies, the main assumption was that the diastolic heart sounds were a stationary process. Since the production of the heart sounds were a stationary process. Since the production of the heart sounds is not a stationary process, a new approach that performs well not only for stationary but also for nonstationary processes can be required. This requirement can be satisfied by the adaptive FTF/FAEST zero tracking method which provides fast and stable convergence as well as computational efficiency since the adaptive FTF/FAEST zero tracking method is based on the exact minimization of least squares criteria and the filter weights of this method are optimal at each time instant. The zero trajectories of the diastolic heart sounds were used to diagnose patients as diseased or normal. Results showed that the normal and abnormal records were incorrectly distinguished in only 6 of 35 cases using a blind protocol where analysis was done without knowledge of the actual disease states of the patients. The most discriminant time region of the zero trajectories of the diastolic heart sounds associated with coronary artery disease was between 200 and 300 msec after the second heart sound during the diastolic period.

Application of adaptive filters to noninvasive acoustical detection of coronary occlusions before and after angioplasty

Previous studies have indicated that coronary stenoses produce sounds due to the turbulent blood flow in these vessels [1]-[10]. Measurement of these signals forms the basis of our noninvasive approach to the detection of coronary artery disease. It is during diastole that coronary blood flow is maximum and the sounds associated with turbulent blood flow through partially occluded coronary arteries would be loudest [1]-[10]. Isolated diastolic heart sounds taken from recordings made at the patient's bedside were modeled using the autoregressive (AR) and autoregressive moving average (ARMA) methods [4], [7] after adaptive line enhancement (ALE). Decisions were made in a blind fashion without prior knowledge of whether a given recording was made before or after angioplasty. Resulting model frequency spectra showed greater high-frequency components (between 400 and 800 Hz) in preangioplasty patients, and a consistent shift in amplitude of the second pole pairs of the AR and ARMA methods with surgery. Blind assessment, based on frequency spectra and poles, correctly classified the diastolic recordings in 18 of 20 cases. These results provide strong evidence supporting our hypothesis that coronary stenoses produce detectable sounds during diastole [1]-[10].

Application of the ARMA method to acoustic detection of coronary artery disease

To further explore the application of advanced signal processing techniques to the noninvasive detection of coronary artery disease, 30 patients (10 angioplasty and 20 normal or abnormal) were tested using autoregressive moving average (ARMA) modelling of the diastolic heart sound data. It is during diastole that coronary blood flow is maximum and sounds associated with turbulent blood flow through partially occluded coronary arteries would be loudest. Model parameters (the power spectral density (PSD) functions and the poles of the ARMA method) were used to separate the normal patients from the abnormal patients in the normal/abnormal study, or to decide whether the recordings were made before or after angioplasty in the angioplasty study. The decisions were made 'blind', without knowledge of the actual disease states of the patients for the normal/abnormal study and without prior knowledge of whether a given recording was made before or after angioplasty for the angioplasty study. Results from the angioplasty and the normal/abnormal studies showed that pre- and post-angioplasty records were correctly distinguished in 8 out of 10 cases, and normal and abnormal records were correctly distinguished in 17 of 20 cases. These results also confirmed that high frequency energy above 400 Hz is probably associated with coronary stenosis.