Performance of an open-source heart sound segmentation algorithm on eight independent databases

OBJECTIVE

Heart sound segmentation is a prerequisite step for the automatic analysis of heart sound signals, facilitating the subsequent identification and classification of pathological events. Recently, hidden Markov model-based algorithms have received increased interest due to their robustness in processing noisy recordings. In this study we aim to evaluate the performance of the recently published logistic regression based hidden semi-Markov model (HSMM) heart sound segmentation method, by using a wider variety of independently acquired data of varying quality.

APPROACH

Firstly, we constructed a systematic evaluation scheme based on a new collection of heart sound databases, which we assembled for the PhysioNet/CinC Challenge 2016. This collection includes a total of more than 120 000 s of heart sounds recorded from 1297 subjects (including both healthy subjects and cardiovascular patients) and comprises eight independent heart sound databases sourced from multiple independent research groups around the world. Then, the HSMM-based segmentation method was evaluated using the assembled eight databases. The common evaluation metrics of sensitivity, specificity, accuracy, as well as the [Formula: see text] measure were used. In addition, the effect of varying the tolerance window for determining a correct segmentation was evaluated.

MAIN RESULTS

The results confirm the high accuracy of the HSMM-based algorithm on a separate test dataset comprised of 102 306 heart sounds. An average [Formula: see text] score of 98.5% for segmenting S1 and systole intervals and 97.2% for segmenting S2 and diastole intervals were observed. The [Formula: see text] score was shown to increases with an increases in the tolerance window size, as expected.

SIGNIFICANCE

The high segmentation accuracy of the HSMM-based algorithm on a large database confirmed the algorithm's effectiveness. The described evaluation framework, combined with the largest collection of open access heart sound data, provides essential resources for evaluators who need to test their algorithms with realistic data and share reproducible results.

Heart sound segmentation using fractal decomposition

In order to assist cardiac diagnosis by phonocardiography, the automated identification of fundamental heart sounds for heart beat segmentation in a cardiac cycle plays a significant role in signal processing. Recent advancements in signal processing have also shown the potential of multifractality in biomedical applications. Hence, in this paper, the multifractal property of heart sounds is utilized to identify first and second heart sounds. The root mean square (rms) fluctuation used to obtain multifractal/singularity spectrum is used to decompose the heart sound into its own fractally-important components in time domain along with simultaneous Gaussianity test to filter out fundamental components. The performance is evaluated on an experimental database of 23 different heart sounds and 6 patients' recordings done in a real clinical environment. Simulation results have shown that it is a promising approach in Heart Sound Segmentation (HSS).

Bidimensional unconstrained optimization approach to EMD: An algorithm revealing skin perfusion alterations in pseudoxanthoma elasticum patients

BACKGROUND AND OBJECTIVE

Pseudoxanthoma elasticum (PXE) is an inherited and systemic metabolic disorder that affects the skin, leading among other things to a peau d'orange appearance. Unfortunately, PXE is still poorly understood and there is no existing therapy to treat the disease. Because the skin is the first organ to be affected in PXE, we propose herein a study of skin microvascular perfusion. By means of this analysis, our goal is to increase knowledge of PXE.

METHODS

For this purpose, microvascular data from patients suffering from PXE and from healthy control subjects were recorded using the laser speckle contrast imaging (LSCI) modality. These data were processed using the recent 2D version of the unconstrained optimization approach to empirical mode decomposition (UOA-EMD). Our work therefore corresponds to the first time this algorithm has been applied to biomedical data.

RESULTS

Our study shows that the 2D-UOA-EMD is able to reveal spatial patterns on local textures of LSCI data. Moreover, these spatial patterns differ between PXE patients and control subjects. Quantification measure of these spatial patterns reveals statistical significant differences between PXE and control subjects, in the neck (p=0.0004) and in the back (p=0.0052).

CONCLUSIONS

For the first time, alterations in microvascular perfusion in PXE patients have been revealed. Our findings open new avenues for our understanding of pathophysiologic skin changes in PXE.

A Noise Reduction Technique Based on Nonlinear Kernel Function for Heart Sound Analysis

The main difficulty encountered in interpretation of cardiac sound is interference of noise. The contaminated noise obscures the relevant information, which are useful for recognition of heart diseases. The unwanted signals are produced mainly by lungs and surrounding environment. In this paper, a novel heart sound denoising technique has been introduced based on a combined framework of wavelet packet transform and singular value decomposition (SVD). The most informative node of the wavelet tree is selected on the criteria of mutual information measurement. Next, the coefficient corresponding to the selected node is processed by the SVD technique to suppress noisy component from heart sound signal. To justify the efficacy of the proposed technique, several experiments have been conducted with heart sound dataset, including normal and pathological cases at different signal to noise ratios. The significance of the method is validated by statistical analysis of the results. The biological information preserved in denoised heart sound signal is evaluated by the k-means clustering algorithm. The overall results show that the proposed method is superior than the baseline methods.

An open access database for the evaluation of heart sound algorithms

In the past few decades, analysis of heart sound signals (i.e. the phonocardiogram or PCG), especially for automated heart sound segmentation and classification, has been widely studied and has been reported to have the potential value to detect pathology accurately in clinical applications. However, comparative analyses of algorithms in the literature have been hindered by the lack of high-quality, rigorously validated, and standardized open databases of heart sound recordings. This paper describes a public heart sound database, assembled for an international competition, the PhysioNet/Computing in Cardiology (CinC) Challenge 2016. The archive comprises nine different heart sound databases sourced from multiple research groups around the world. It includes 2435 heart sound recordings in total collected from 1297 healthy subjects and patients with a variety of conditions, including heart valve disease and coronary artery disease. The recordings were collected from a variety of clinical or nonclinical (such as in-home visits) environments and equipment. The length of recording varied from several seconds to several minutes. This article reports detailed information about the subjects/patients including demographics (number, age, gender), recordings (number, location, state and time length), associated synchronously recorded signals, sampling frequency and sensor type used. We also provide a brief summary of the commonly used heart sound segmentation and classification methods, including open source code provided concurrently for the Challenge. A description of the PhysioNet/CinC Challenge 2016, including the main aims, the training and test sets, the hand corrected annotations for different heart sound states, the scoring mechanism, and associated open source code are provided. In addition, several potential benefits from the public heart sound database are discussed.

Multistage decision-based heart sound delineation method for automated analysis of heart sounds and murmurs

A robust multistage decision-based heart sound delineation (MDHSD) method is presented for automatically determining the boundaries and peaks of heart sounds (S1, S2, S3, and S4), systolic, and diastolic murmurs (early, mid, and late) and high-pitched sounds (HPSs) of the phonocardiogram (PCG) signal. The proposed MDHSD method consists of the Gaussian kernels based signal decomposition (GSDs) and multistage decision-based delineation (MDBD). The GSD algorithm first removes the low-frequency (LF) artefacts and then decomposes the filtered signal into two subsignals: the LF sound part (S1, S2, S3, and S4) and the high-frequency sound part (murmurs and HPSs). The MDBD algorithm consists of absolute envelope extraction, adaptive thresholding, and fiducial point determination. The accuracy and robustness of the proposed method is evaluated using various types of normal and pathological PCG signals. Results show that the method achieves an average sensitivity of 98.22%, positive predictivity of 97.46%, and overall accuracy of 95.78%. The method yields maximum average delineation errors of 4.52 and 4.14 ms for determining the start-point and end-point of sounds. The proposed multistage delineation algorithm is capable of improving the delineation accuracy under time-varying amplitudes of heart sounds and various types of murmurs. The proposed method has significant potential applications in heart sounds and murmurs classification systems.

Logistic Regression-HSMM-Based Heart Sound Segmentation

The identification of the exact positions of the first and second heart sounds within a phonocardiogram (PCG), or heart sound segmentation, is an essential step in the automatic analysis of heart sound recordings, allowing for the classification of pathological events. While threshold-based segmentation methods have shown modest success, probabilistic models, such as hidden Markov models, have recently been shown to surpass the capabilities of previous methods. Segmentation performance is further improved when a priori information about the expected duration of the states is incorporated into the model, such as in a hidden semi-Markov model (HSMM). This paper addresses the problem of the accurate segmentation of the first and second heart sound within noisy real-world PCG recordings using an HSMM, extended with the use of logistic regression for emission probability estimation. In addition, we implement a modified Viterbi algorithm for decoding the most likely sequence of states, and evaluated this method on a large dataset of 10,172 s of PCG recorded from 112 patients (including 12,181 first and 11,627 second heart sounds). The proposed method achieved an average F1 score of 95.63 ± 0.85%, while the current state of the art achieved 86.28 ± 1.55% when evaluated on unseen test recordings. The greater discrimination between states afforded using logistic regression as opposed to the previous Gaussian distribution-based emission probability estimation as well as the use of an extended Viterbi algorithm allows this method to significantly outperform the current state-of-the-art method based on a two-sided paired t-test.

The electronic stethoscope

Most heart diseases are associated with and reflected by the sounds that the heart produces. Heart auscultation, defined as listening to the heart sound, has been a very important method for the early diagnosis of cardiac dysfunction. Traditional auscultation requires substantial clinical experience and good listening skills. The emergence of the electronic stethoscope has paved the way for a new field of computer-aided auscultation. This article provides an in-depth study of (1) the electronic stethoscope technology, and (2) the methodology for diagnosis of cardiac disorders based on computer-aided auscultation. The paper is based on a comprehensive review of (1) literature articles, (2) market (state-of-the-art) products, and (3) smartphone stethoscope apps. It covers in depth every key component of the computer-aided system with electronic stethoscope, from sensor design, front-end circuitry, denoising algorithm, heart sound segmentation, to the final machine learning techniques. Our intent is to provide an informative and illustrative presentation of the electronic stethoscope, which is valuable and beneficial to academics, researchers and engineers in the technical field, as well as to medical professionals to facilitate its use clinically. The paper provides the technological and medical basis for the development and commercialization of a real-time integrated heart sound detection, acquisition and quantification system.

Automatic Differentiation of Normal and Continuous Adventitious Respiratory Sounds Using Ensemble Empirical Mode Decomposition and Instantaneous Frequency

Differentiating normal from adventitious respiratory sounds (RS) is a major challenge in the diagnosis of pulmonary diseases. Particularly, continuous adventitious sounds (CAS) are of clinical interest because they reflect the severity of certain diseases. This study presents a new classifier that automatically distinguishes normal sounds from CAS. It is based on the multiscale analysis of instantaneous frequency (IF) and envelope (IE) calculated after ensemble empirical mode decomposition (EEMD). These techniques have two major advantages over previous techniques: high temporal resolution is achieved by calculating IF-IE and a priori knowledge of signal characteristics is not required for EEMD. The classifier is based on the fact that the IF dispersion of RS signals markedly decreases when CAS appear in respiratory cycles. Therefore, CAS were detected by using a moving window to calculate the dispersion of IF sequences. The study dataset contained 1494 RS segments extracted from 870 inspiratory cycles recorded from 30 patients with asthma. All cycles and their RS segments were previously classified as containing normal sounds or CAS by a highly experienced physician to obtain a gold standard classification. A support vector machine classifier was trained and tested using an iterative procedure in which the dataset was randomly divided into training (65%) and testing (35%) sets inside a loop. The SVM classifier was also tested on 4592 simulated CAS cycles. High total accuracy was obtained with both recorded (94.6% ± 0.3%) and simulated (92.8% ± 3.6%) signals. We conclude that the proposed method is promising for RS analysis and classification.