In this letter, we propose a method for the automated detection of heart valve disorders namely, the aortic stenosis (AS), mitral stenosis (MS), and mitral regurgitation (MR) from the phonocardiogram (PCG) signal. The wavelet synchrosqueezing transform is used to obtain the time-frequency matrix from the segmented cycles of the PCG signal. From the time-frequency matrix, the magnitude and phase features are extracted. The random forest (RF) classifier is used for the classification. The results reveal that the proposed method has the average individual accuracy (IA) values of 98.83%, 97.66%, 91.16%, and 92.83% for normal, AS, MS, and MR classes.

Automated Detection of Heart Valve Disorders From the PCG Signal Using Time-Frequency Magnitude and Phase Features

According to the World Health Organization, Coronary Artery Disease (CAD) is a leading cause of death globally. CAD is categorized into three types, namely Single Vessel Coronary Artery Disease (SVCAD), Double Vessel Coronary Artery Disease (DVCAD), and Triple Vessel Coronary Artery Disease (TVCAD). At present, angiography is the most popular technique to detect CAD that is quite expensive and invasive. Phonocardiogram (PCG), being economical and non-invasive, is a crucial modality towards the detection of cardiac disorders, but only trained medical professionals can interpret heart auscultations in clinical environments. This research aims to detect CAD and its types from PCG signatures through feature fusion and a two-stage classification strategy. The self-developed low-cost stethoscope was used to collect PCG data from a local hospital. The PCG signals were preprocessed through an iterative signal decomposition method known as Empirical Mode Decomposition (EMD). EMD decomposes the raw PCG signal into its constituent components called Intrinsic Mode Functions (IMFs). Preprocessed PCG signal was generated exclusively through combining those signal components that contain high discriminative characteristics and less redundancy. Next, Mel Frequency Cepstral Coefficients (MFCCs), spectral and statistical features were extracted. A two-stage classification framework was devised to identify healthy and CAD types. The first stage framework relies on the fusion of MFCC and statistical features with the K-nearest neighbor classifier to predict normal and CAD cases. The second stage is activated only when the first stage detects CAD. The fusion of spectral, statistical, and MFCC features was employed with Support Vector Machines classifier to categorize PCG signatures into DVCAD, SVCAD, and TVCAD classes in the second stage. The proposed method yields mean accuracy values of 88.0%, 89.2%, 91.1%, and 85.3% for normal, DVCAD, SVCAD, and TVCAD, respectively, through 10-fold cross-validation. Comparative analysis with existing approaches confirmed the reliability of the proposed method for categorizing CAD in general clinical environments. The proposed model enhances the diagnosis performance by providing a second opinion during the medical examination.

A two-stage classification model integrating feature fusion for coronary artery disease detection and classification

A fusion framework based on multi-domain features and deep learning features of phonocardiogram for coronary artery disease detection.

Application of Petersen graph pattern technique for automated detection of heart valve diseases with PCG signals

Novel three kernelled binary pattern feature extractor based automated PCG sound classification method

Classification of coronary artery diseased and normal subjects using multi-channel phonocardiogram signal

Detection of coronary artery atherosclerotic disease using novel features from synchrosqueezing transform of phonocardiogram

An improved method to detect coronary artery disease using phonocardiogram signals in noisy environment

Heart sounds provide important information about cardiac status either through auscultation or by processing of electronic recordings of heart sounds, termed as phonocardiogram (PCG). The current study analyzes the relation between PCG signal from four auscultation sites for diagnosis of coronary artery disease (CAD). For this purpose, data collected from four locations on chest is analysed using cross power spectral density (CPSD). The features from spectrum are derived using distribution of power and their moments in subbands. ReliefF algorithm is employed for selecting features to be used in classification framework. Performance is evaluated using five-fold cross-validation in a support vector machine (SVM) classifier. Experimental results show that multichannel analysis performs better than existing features, as well as for same CPSD based features derived from single channel power spectrum. Different subband width were experimentally analysed to find an optimal width for feature extraction. The study shows the potential of using connectivity between PCG signals from multiple sites for diagnosing CAD related abnormality.

Identification of Coronary Artery Disease using Cross Power Spectral Density

Coronary artery disease (CAD) is one of the leading causes of mortality and morbidity globally. Nowadays, it is spreading at an alarming rate. Recently, there is an increasing interest to develop simple and non-invasive automated methods for reliable diagnosis of CAD. Studies reported that the use of single-channel phonocardiogram (PCG) signal for detecting weak CAD murmurs caused by the stenosed coronary arteries due to turbulent blood flow. In this work, we introduce a new framework with multi-channel data acquisition system to classify CAD and normal subjects. The proposed method does not require any reference signal such as an electrocardiogram (ECG) signal for PCG signal segmentation as reported in the earlier studies. Subsequently, the study has used five different features, such as spectral moments, spectral entropy, moments of PSD function, autoregressive (AR) parameters, and instantaneous frequency derived from frequency representations of PCG signals. These features have captured the specific details related to the disease. We use an artificial neural network (ANN) for the classification task. Experimental results show that the AR features well-performed. We achieve an accuracy of 74.24% by using multi-channel recorded data where as the best performance obtained using single-channel signal is 69.69%.

Kayapanda Mandana

Papers

Classification of coronary artery diseased and normal subjects using multi-channel phonocardiogram signal

Detection of coronary artery atherosclerotic disease using novel features from synchrosqueezing transform of phonocardiogram

An improved method to detect coronary artery disease using phonocardiogram signals in noisy environment

Identification of Coronary Artery Disease using Cross Power Spectral Density

Identification of Coronary Artery Diseased Subjects Using Spectral Featuries