Computer-aided auscultation of the heart: from anatomy and physiology to diagnostic decision support

An automatic approach for heart failure typing based on heart sounds and convolutional recurrent neural networks

Heart failure (HF) is a complex clinical syndrome that poses a major hazard to human health. Patients with different types of HF have great differences in pathogenesis and treatment options. Therefore, HF typing is of great significance for timely treatment of patients. In this paper, we proposed an automatic approach for HF typing based on heart sounds (HS) and convolutional recurrent neural networks, which provides a new non-invasive and convenient way for HF typing. Firstly, the collected HS signals were preprocessed with adaptive wavelet denoising. Then, the logistic regression based hidden semi-Markov model was utilized to segment HS frames. For the distinction between normal subjects and the HF patients with preserved ejection fraction or reduced ejection fraction, a model based on convolutional neural network and recurrent neural network was built. The model can automatically learn the spatial and temporal characteristics of HS signals. The results show that the proposed model achieved a superior performance with an accuracy of 97.64%. This study suggests the proposed method could be a useful tool for HF recognition and as a supplement for HF typing.

A Review of Computer-Aided Heart Sound Detection Techniques

Cardiovascular diseases have become one of the most prevalent threats to human health throughout the world. As a noninvasive assistant diagnostic tool, the heart sound detection techniques play an important role in the prediction of cardiovascular diseases. In this paper, the latest development of the computer-aided heart sound detection techniques over the last five years has been reviewed. There are mainly the following aspects: the theories of heart sounds and the relationship between heart sounds and cardiovascular diseases; the key technologies used in the processing and analysis of heart sound signals, including denoising, segmentation, feature extraction and classification; with emphasis, the applications of deep learning algorithm in heart sound processing. In the end, some areas for future research in computer-aided heart sound detection techniques are explored, hoping to provide reference to the prediction of cardiovascular diseases.

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.

Cardiac Auscultation in the Modern Era: Premature Requiem or Phoenix Rising?

Competent cardiac auscultation remains a most important skill for the detection of heart disease. Currently it is poorly taught and often ignored or poorly performed, resulting in inaccurate and inefficient patient assessments. This review documents that teaching can be over 90% effective with new, proven teaching methods emphasizing repetition and normal-abnormal comparisons of sounds, using computer-aided and online resources. At present, these concepts are not widely adopted by medical schools. Our current knowledge of teaching heart auscultation is critically reviewed, including traditional bedside, clinic and classroom settings, as well as computer, simulator, and multimedia-based learning. The assessment of auscultation skill in the learning process. The adoption of competence-based learning promises to integrate the assessment of auscultation skill in the learning process. Newer teaching methods, such as auditory training and repetitive listening, offer excellent murmur recognition and diagnosis learning, and hand-held ultrasound is proposed as a helpful adjunct to teaching auscultation. Although ongoing research remains important to develop better teaching methods, the adoption of proven existing concepts has great potential to improve teaching and practice of this valuable skill.

Automatic heart sound segmentation and murmur detection in pediatric phonocardiograms

The digital analysis of heart sounds has revealed itself as an evolving field of study. In recent years, numerous approaches to create decision support systems were attempted. This paper proposes two novel algorithms: one for the segmentation of heart sounds into heart cycles and another for detecting heart murmurs. The segmentation algorithm, based on the autocorrelation function to find the periodic components of the PCG signal had a sensitivity and positive predictive value of 89.2% and 98.6%, respectively. The murmur detection algorithm is based on features collected from different domains and was evaluated in two ways: a random division between train and test set and a division according to patients. The first returned sensitivity and specificity of 98.42% and 97.21% respectively for a minimum error of 2.19%. The second division had a far worse performance with a minimum error of 33.65%. The operating point was chosen at sensitivity 69.67% and a specificity 46.91% for a total error of 38.90% by varying the percentage of segments classified as murmurs needed for a positive murmur classification.

A Novel Method for Screening Children with Isolated Bicuspid Aortic Valve

This paper presents a novel processing method for heart sound signal: the statistical time growing neural network (STGNN). The STGNN performs a robust classification by merging supervised and unsupervised statistical methods to overcome non-stationary behavior of the signal. By combining available preprocessing and segmentation techniques and the STGNN classifier, we build an automatic tool for screening children with isolated BAV, the congenital heart malformation which can lead to serious cardiovascular lesions. Children with BAV (22 individuals) and healthy condition (28 individuals) are subjected to the study. The performance of the STGNN is compared to that of a time growing neural network (CTGNN) and a conventional support vector (CSVM) machine, using balanced repeated random sub sampling. The average of the accuracy/sensitivity is estimated to be 87.4/86.5 for the STGNN, 81.8/83.4 for the CTGNN, and 72.9/66.8 for the CSVM. Results show that the STGNN offers better performance and provides more immunity to the background noise as compared to the CTGNN and CSVM. The method is implementable in a computer system to be employed in primary healthcare centers to improve the screening accuracy.

Multi-point accelerometric detection and principal component analysis of heart sounds

Heart sounds are a fundamental physiological variable that provide a unique insight into cardiac semiotics. However a deterministic and unambiguous association between noises in cardiac dynamics is far from being accomplished yet due to many and different overlapping events which contribute to the acoustic emission. The current computer-based capacities in terms of signal detection and processing allow one to move from the standard cardiac auscultation, even in its improved forms like electronic stethoscopes or hi-tech phonocardiography, to the extraction of information on the cardiac activity previously unexplored. In this report, we present a new equipment for the detection of heart sounds, based on a set of accelerometric sensors placed in contact with the chest skin on the precordial area, and are able to measure simultaneously the vibration induced on the chest surface by the heart's mechanical activity. By utilizing advanced algorithms for the data treatment, such as wavelet decomposition and principal component analysis, we are able to condense the spatially extended acoustic information and to provide a synthetical representation of the heart activity. We applied our approach to 30 adults, mixed per gender, age and healthiness, and correlated our results with standard echocardiographic examinations. We obtained a 93% concordance rate with echocardiography between healthy and unhealthy hearts, including minor abnormalities such as mitral valve prolapse.

Biomedical informatics for computer-aided decision support systems: a survey

The volumes of current patient data as well as their complexity make clinical decision making more challenging than ever for physicians and other care givers. This situation calls for the use of biomedical informatics methods to process data and form recommendations and/or predictions to assist such decision makers. The design, implementation, and use of biomedical informatics systems in the form of computer-aided decision support have become essential and widely used over the last two decades. This paper provides a brief review of such systems, their application protocols and methodologies, and the future challenges and directions they suggest.

Robust heart rate measurement with phonocardiogram by on-line template extraction and matching

As health care becomes popular, daily monitoring of health-status related parameters, including the heart rate (HR), is getting more and more valued. An easy, comfortable and robust solution is therefore an important issue. Phonocardiogram (PCG) is a physiological signal reflecting the cardiovascular status. It could be recorded by microphone-equipped on-hand devices, like the smartphone, even without direct skin contact. However, high inter- and intra-variance of PCG make its processing challenging. For PCG-based HR measurement, a robust method is still strongly required. In this paper, we propose a HR measurement algorithm on the processing of PCG that uses on-line template extraction and matching. Through several experiments where traditional methods cannot effectively handle, the robustness of our method is verified by its accurate HR measurement results.

Development of a pediatric cardiac computer aided auscultation decision support system

Developing countries have a large population of children living with undiagnosed heart murmurs. As a result of an accompanying skills shortage, most of these children will not get the necessary treatment. The objective of this paper was to develop a decision support system. This could enable health care providers in developing countries with tools to screen large amounts of children without the need for expensive equipment or specialist skills. For this purpose an algorithm was designed and tested to detect heart murmurs in digitally recorded signals. A specificity of 94% and a sensitivity of 91% were achieved using novel signal processing techniques and an ensemble of neural networks as classifier.

Modulation filtering for noise detection in heart sound signals

Cardiac auscultation has proven to be an excellent diagnostic tool. Heart sound processing algorithms are not completely robust in the presence of noise, requiring clean segments of heart sounds to extract reliable diagnostic features. This paper presents a new approach to detect transient noises mixed with heart sound. The algorithm explores a single channel source separation algorithm and evaluates the non-stationary separated signals. It has the potential to be applied in real-time. Using a database of heart sounds acquired in real-life scenario, the method showed a sensitivity and a specificity of 93.6% and 92.3%, respectively.