Detection of heart murmurs using wavelet analysis and artificial neural networks

This paper presents the algorithm and technical aspects of an intelligent diagnostic system for the detection of heart murmurs. The purpose of this research is to address the lack of effectively accurate cardiac auscultation present at the primary care physician office by development of an algorithm capable of operating within the hectic environment of the primary care office. The proposed algorithm consists of three main stages. First; denoising of input data (digital recordings of heart sounds), via Wavelet Packet Analysis. Second; input vector preparation through the use of Principal Component Analysis and block processing. Third; classification of the heart sound using an Artificial Neural Network. Initial testing revealed the intelligent diagnostic system can differentiate between normal healthy heart sounds and abnormal heart sounds (e.g., murmurs), with a specificity of 70.5% and a sensitivity of 64.7%.     

PHONOCARDIOGRAPHY

In phonocardiography the second heart sound is important in appraisal of congenital heart disease and pulmonary hypertension because it reflects the duration of right ventricular systoles.The systolic murmur in patients with intracardiac shunt decreases as pulmonary hypertension develops, and may eventually disappear completely as the pulmonary pressure reaches systemic level.Reference tracings in phonocardiography are useful for showing the interrelationship of the various cardiac sounds and murmurs.     

Hemodynamics-driven mathematical model of first and second heart sound generation

We propose a novel, two-degree of freedom mathematical model of mechanical vibrations of the heart that generates heart sounds in CircAdapt, a complete real-time model of the cardiovascular system. Heart sounds during rest, exercise, biventricular (BiVHF), left ventricular (LVHF) and right ventricular heart failure (RVHF) were simulated to examine model functionality in various conditions. Simulated and experimental heart sound components showed both qualitative and quantitative agreements in terms of heart sound morphology, frequency, and timing. Rate of left ventricular pressure (LV dp/dt_max) and first heart sound (S1) amplitude were proportional with exercise level. The relation of the second heart sound (S2) amplitude with exercise level was less significant. BiVHF resulted in amplitude reduction of S1. LVHF resulted in reverse splitting of S2 and an amplitude reduction of only the left-sided heart sound components, whereas RVHF resulted in a prolonged splitting of S2 and only a mild amplitude reduction of the right-sided heart sound components. In conclusion, our hemodynamics-driven mathematical model provides fast and realistic simulations of heart sounds under various conditions and may be helpful to find new indicators for diagnosis and prognosis of cardiac diseases.New & noteworthyTo the best of our knowledge, this is the first hemodynamic-based heart sound generation model embedded in a complete real-time computational model of the cardiovascular system. Simulated heart sounds are similar to experimental and clinical measurements, both quantitatively and qualitatively. Our model can be used to investigate the relationships between heart sound acoustic features and hemodynamic factors/anatomical parameters.     

Cluster analysis and classification of heart sounds

Acoustic heart signals, generated by the mechanical processes of the cardiac cycle, carry significant information about the underlying functioning of the cardiovascular system. We describe a computational analysis framework for identifying distinct morphologies of heart sounds and classifying them into physiological states. The analysis framework is based on hierarchical clustering, compact data representation in the feature space of cluster distances and a classification algorithm. We applied the proposed framework on two heart sound datasets, acquired during controlled alternations of the physiological conditions, and analyzed the morphological changes induced to the first heart sound (S1), and the ability to predict physiological variables from the morphology of S1. On the first dataset of 12 subjects, acquired while modulating the respiratory pressure, the algorithm achieved an average accuracy of 82 ± 7% in classifying the level of breathing resistance, and was able to estimate the instantaneous breathing pressure with an average error of 19 ± 6%. A strong correlation of 0.92 was obtained between the estimated and the actual breathing efforts. On the second dataset of 11 subjects, acquired during pharmacological stress tests, the average accuracy in classifying the stress stage was 86 ± 7%. The effects of the chosen raw signal representation, distance metrics and classification algorithm on the performance were studied on both real and simulated data. The results suggest that quantitative heart sound analysis may provide a new non-invasive technique for continuous cardiac monitoring and improved detection of mechanical dysfunctions caused by cardiovascular and cardiopulmonary diseases.     

The Human Brain Maintains Contradictory and Redundant Auditory Sensory Predictions

Computational and experimental research has revealed that auditory sensory predictions are derived from regularities of the current environment by using internal generative models. However, so far, what has not been addressed is how the auditory system handles situations giving rise to redundant or even contradictory predictions derived from different sources of information. To this end, we measured error signals in the event-related brain potentials (ERPs) in response to violations of auditory predictions. Sounds could be predicted on the basis of overall probability, i.e., one sound was presented frequently and another sound rarely. Furthermore, each sound was predicted by an informative visual cue. Participants’ task was to use the cue and to discriminate the two sounds as fast as possible. Violations of the probability based prediction (i.e., a rare sound) as well as violations of the visual-auditory prediction (i.e., an incongruent sound) elicited error signals in the ERPs (Mismatch Negativity [MMN] and Incongruency Response [IR]). Particular error signals were observed even in case the overall probability and the visual symbol predicted different sounds. That is, the auditory system concurrently maintains and tests contradictory predictions. Moreover, if the same sound was predicted, we observed an additive error signal (scalp potential and primary current density) equaling the sum of the specific error signals. Thus, the auditory system maintains and tolerates functionally independently represented redundant and contradictory predictions. We argue that the auditory system exploits all currently active regularities in order to optimally prepare for future events.     

Discovery of multiple level heart-sound morphological variability resulting from changes in physiological states

Heart sounds carry information about the mechanical activity of the cardiovascular system. This information includes the specific physiological state of the subject, and short term variability related to the respiratory cycle. The interpretation of the sounds and extraction of changes in the physiological state, while monitoring short term variability is still an open problem and is the subject of this paper.We present a novel computational framework for analysis of data with multi-level variability, caused by externally induced changes. The framework presented includes an initial clustering of the first heart sound (S1) according to the morphology, and further aggregation of clusters into super-clusters. The clusters and super clusters are two methods of data segmentation, each reflecting a different level of variability in the data.The framework is applied to heart sounds recorded during laparoscopic surgeries of six patients. Procedures of this kind include anesthesia and abdominal insufflation, which together with the respiratory cycle, induce changes to the heart sound signal. We demonstrate a separation of the heart sound morphology according to different physiological states. The physiological states considered are the respiratory cycle, and the stages of the surgery. We achieve results of 90 ± 4% classification accuracy of heart beats to operation stages.The proposed framework is general and can be used to analyze data characterized by multi-level variability for various other (biomedical) applications.     

Segmentation of heart sounds based on dynamic clustering

The heart sound signal is first separated into cycles, where the cycle detection is based on an instantaneous cycle frequency. The heart sound data of one cardiac cycle can be decomposed into a number of atoms characterized by timing delay, frequency, amplitude, time width and phase. To segment heart sounds, we made a hypothesis that the atoms of a heart sound congregate as a cluster in time–frequency domains. We propose an atom density function to indicate clusters. To suppress clusters of murmurs and noise, weighted density function by atom energy is further proposed to improve the segmentation of heart sounds. Therefore, heart sounds are indicated by the hybrid analysis of clustering and medical knowledge. The segmentation scheme is automatic and no reference signal is needed. Twenty-six subjects, including 3 normal and 23 abnormal subjects, were tested for heart sound signals in various clinical cases. Our statistics show that the segmentation was successful for signals collected from normal subjects and patients with moderate murmurs.     

Respiratory sound analysis in healthy and pathological subjects: A wavelet approach

In this paper we describe the application of a wavelet analysis-based method, to characterize the frequency power distribution of the unsteady respiratory sound signals in order to better discriminate the healthy state of a given subject. To evaluate the methodology, both normal tracheal sounds as well as adventitious respiratory sounds were investigated. In particular, our analysis shows the possibility to extract useful statistical information on the energy content and its mean frequency distribution giving us a quantitative characteristic hallmark of the respiratory pattern. The presence of sound anomalies can be pointed out through some specific patterns of the wavelet mean power spectra and thus the localization of the related quartiles which can be used as simple and efficient diagnostic indices. In this study the method has been applied in healthy subjects and patients with different respiratory diseases. Results show that different power spectra patterns characterize health from disease. Some preliminary results indicate also that pathological patterns can change as result of therapeutical interventions like mechanical ventilation.     

Complex networks and deep learning for EEG signal analysis

Electroencephalogram (EEG) signals acquired from brain can provide an effective representation of the human’s physiological and pathological states. Up to now, much work has been conducted to study and analyze the EEG signals, aiming at spying the current states or the evolution characteristics of the complex brain system. Considering the complex interactions between different structural and functional brain regions, brain network has received a lot of attention and has made great progress in brain mechanism research. In addition, characterized by autonomous, multi-layer and diversified feature extraction, deep learning has provided an effective and feasible solution for solving complex classification problems in many fields, including brain state research. Both of them show strong ability in EEG signal analysis, but the combination of these two theories to solve the difficult classification problems based on EEG signals is still in its infancy. We here review the application of these two theories in EEG signal research, mainly involving brain–computer interface, neurological disorders and cognitive analysis. Furthermore, we also develop a framework combining recurrence plots and convolutional neural network to achieve fatigue driving recognition. The results demonstrate that complex networks and deep learning can effectively implement functional complementarity for better feature extraction and classification, especially in EEG signal analysis.     

An algorithm for FHR estimation from foetal phonocardiographic signals

The long-term foetal surveillance is often to be recommended. Hence, the fully non-invasive acoustic recording, through maternal abdomen, represents a valuable alternative to the ultrasonic cardiotocography. Unfortunately, the recorded heart sound signal is heavily loaded by noise, thus the determination of the foetal heart rate raises serious signal processing issues. In this paper, we present a new algorithm for foetal heart rate estimation from foetal phonocardiographic recordings. A filtering is employed as a first step of the algorithm to reduce the background noise. A block for first heart sounds enhancing is then used to further reduce other components of foetal heart sound signals. A complex logic block, guided by a number of rules concerning foetal heart beat regularity, is proposed as a successive block, for the detection of most probable first heart sounds from several candidates. A final block is used for exact first heart sound timing and in turn foetal heart rate estimation. Filtering and enhancing blocks are actually implemented by means of different techniques, so that different processing paths are proposed. Furthermore, a reliability index is introduced to quantify the consistency of the estimated foetal heart rate and, based on statistic parameters; [,] a software quality index is designed to indicate the most reliable analysis procedure (that is, combining the best processing path and the most accurate time mark of the first heart sound, provides the lowest estimation errors). The algorithm performances have been tested on phonocardiographic signals recorded in a local gynaecology private practice from a sample group of about 50 pregnant women. Phonocardiographic signals have been recorded simultaneously to ultrasonic cardiotocographic signals in order to compare the two foetal heart rate series (the one estimated by our algorithm and the other provided by cardiotocographic device). Our results show that the proposed algorithm, in particular some analysis procedures, provides reliable foetal heart rate signals, very close to the reference cardiotocographic recordings.     

Machine learning approach for automatic recognition of tomato-pollinating bees based on their buzzing-sounds

Bee-mediated pollination greatly increases the size and weight of tomato fruits. Therefore, distinguishing between the local set of bees–those that are efficient pollinators–is essential to improve the economic returns for farmers. To achieve this, it is important to know the identity of the visiting bees. Nevertheless, the traditional taxonomic identification of bees is not an easy task, requiring the participation of experts and the use of specialized equipment. Due to these limitations, the development and implementation of new technologies for the automatic recognition of bees become relevant. Hence, we aim to verify the capacity of Machine Learning (ML) algorithms in recognizing the taxonomic identity of visiting bees to tomato flowers based on the characteristics of their buzzing sounds. We compared the performance of the ML algorithms combined with the Mel Frequency Cepstral Coefficients (MFCC) and with classifications based solely on the fundamental frequency, leading to a direct comparison between the two approaches. In fact, some classifiers powered by the MFCC–especially the SVM–achieved better performance compared to the randomized and sound frequency-based trials. Moreover, the buzzing sounds produced during sonication were more relevant for the taxonomic recognition of bee species than analysis based on flight sounds alone. On the other hand, the ML classifiers performed better in recognizing bees genera based on flight sounds. Despite that, the maximum accuracy obtained here (73.39% by SVM) is still low compared to ML standards. Further studies analyzing larger recording samples, and applying unsupervised learning systems may yield better classification performance. Therefore, ML techniques could be used to automate the taxonomic recognition of flower-visiting bees of the cultivated tomato and other buzz-pollinated crops. This would be an interesting option for farmers and other professionals who have no experience in bee taxonomy but are interested in improving crop yields by increasing pollination.     

Heart energy signature spectrogram for cardiovascular diagnosis

A new method and application is proposed to characterize intensity and pitch of human heart sounds and murmurs. Using recorded heart sounds from the library of one of the authors, a visual map of heart sound energy was established. Both normal and abnormal heart sound recordings were studied. Representation is based on Wigner-Ville joint time-frequency transformations. The proposed methodology separates acoustic contributions of cardiac events simultaneously in pitch, time and energy. The resolution accuracy is superior to any other existing spectrogram method. The characteristic energy signature of the innocent heart murmur in a child with the S3 sound is presented. It allows clear detection of S1, S2 and S3 sounds, S2 split, systolic murmur, and intensity of these components. The original signal, heart sound power change with time, time-averaged frequency, energy density spectra and instantaneous variations of power and frequency/pitch with time, are presented. These data allow full quantitative characterization of heart sounds and murmurs. High accuracy in both time and pitch resolution is demonstrated. Resulting visual images have self-referencing quality, whereby individual features and their changes become immediately obvious.     

A Humanoid Method for Extracting Abnormal Engine Sounds from Engine Acoustics Based on Adaptive Voiterra Filter

The improvement of SNR (Signal-to-Noise Ratio) of abnormal engine sounds is of great help in improving the accuracy of engine fault diagnosis.By imitating the way that human technicians use to distinguish abnormal engine sounds from engine acoustics,a humanoid abnormal sound extracting method is proposed.By implementing adaptive Volterra filter in the canonical Adaptive Noise Cancellation (ANC) system,the proposed method is capable of tracing the engine baseline sound which exhibits an intrinsic nonlinear dynamics.Besides,by introducing a template noise tailored from the records of engine baseline sound and taking it as virtual input of the adaptive Volterra filter,the priori knowledge of engine baseline sound,such as inherent correlation,periodicity or phase information,and stochastic factors,is taken into consideration.The hybrid simulations prove that the proposed method is functional.Since the method proposed is essentially a single-sensor based ANC,hopefully,it may become an effective way to extricate the dilemma that canonical dual-sensor based ANC encounters when it is used in extracting fault-featured signals from observed signals.     

Discovering Associations in Biomedical Datasets by Link-based Associative Classifier (LAC)

Associative classification mining (ACM) can be used to provide predictive models with high accuracy as well as interpretability. However, traditional ACM ignores the difference of significances among the features used for mining. Although weighted associative classification mining (WACM) addresses this issue by assigning different weights to features, most implementations can only be utilized when pre-assigned weights are available. In this paper, we propose a link-based approach to automatically derive weight information from a dataset using link-based models which treat the dataset as a bipartite model. By combining this link-based feature weighting method with a traditional ACM method–classification based on associations (CBA), a Link-based Associative Classifier (LAC) is developed. We then demonstrate the application of LAC to biomedical datasets for association discovery between chemical compounds and bioactivities or diseases. The results indicate that the novel link-based weighting method is comparable to support vector machine (SVM) and RELIEF method, and is capable of capturing significant features. Additionally, LAC is shown to produce models with high accuracies and discover interesting associations which may otherwise remain unrevealed by traditional ACM.     

A robust heart sounds segmentation module based on S-transform

This paper presents a new module for heart sounds segmentation based on S-transform. The heart sounds segmentation process segments the PhonoCardioGram (PCG) signal into four parts: S1 (first heart sound), systole, S2 (second heart sound) and diastole. It can be considered one of the most important phases in the auto-analysis of PCG signals. The proposed segmentation module can be divided into three main blocks: localization of heart sounds, boundaries detection of the localized heart sounds and classification block to distinguish between S1 and S2. An original localization method of heart sounds are proposed in this study. The method named SSE calculates the Shannon energy of the local spectrum calculated by the S-transform for each sample of the heart sound signal. The second block contains a novel approach for the boundaries detection of S1 and S2. The energy concentrations of the S-transform of localized sounds are optimized by using a window width optimization algorithm. Then the SSE envelope is recalculated and a local adaptive threshold is applied to refine the estimated boundaries. To distinguish between S1 and S2, a feature extraction method based on the singular value decomposition (SVD) of the S-matrix is applied in this study. The proposed segmentation module is evaluated at each block according to a database of 80 sounds, including 40 sounds with cardiac pathologies.     

Neurogenesis or neuronal specification: phosphorylation strikes again!

One way to localize sounds is to measure differences in sound intensity at the two ears. This comparison is made in the lateral superior olive, where signals from both ears converge. Magnusson et al. in this issue of Neuron show that dendritic GABA release can regulate this comparison, which may allow animals localizing sounds to adapt to listening conditions.     

Acoustic signalling in female fish: factors influencing sound characteristics in croaking gouramis

The characteristics of sounds produced by fishes are influenced by several factors such as size. The current study analyses factors affecting structural properties of acoustic signals produced by female croaking gouramis Trichopsis vittata during agonistic interactions. Female sounds (although seldom analysed separately from male sounds) can equally be used to investigate factors affecting the sound characteristics in fish. Sound structure, dominant frequency and sound pressure levels (SPL) were determined and correlated to body size and the order in which sounds were emitted. Croaking sounds consisted of series of single-pulsed or double-pulsed bursts, each burst produced by one pectoral fin. Main energies were concentrated between 1.3 and 1.5 kHz. The dominant frequency decreased with size, as did the percentage of single-pulsed bursts within croaking sounds. The SPL and the number of bursts within a sound were independent of size but decreased significantly with the order of their production. Thus, acoustic signals produced at the beginning of agonistic interactions were louder and consisted of more bursts than subsequent ones. Our data indicate that body size affects the dominant frequency and structure of sounds. The increase in the percentage of double-pulsed bursts with size may be due to stronger pectoral muscles in larger fish. In contrast, ongoing fights apparently result in muscle fatigue and subsequently in a decline in the number of bursts and SPL. The factor ‘order of sound production’ points to an intra-individual variability of sounds and should be considered in future studies.     

Phonographic diagnostic aid in heart defects using neural networks

New approaches to pattern recognition and modeling using neural networks also provide powerful tools for the analysis of heart sounds in the diagnosis of heart failure. This paper describes how a neutral network can be used to estimate the duration of systolic and diastolic heart phases as well as to suggest a suitable diagnosis. A neural network-based auscultation system is capable of documenting and analysing a heart sound signal by applying methods similar to those used by physicians in the subjective interpretation of stethoscopic signals.     

Pathological cardiac remodeling seen by the eyes of proteomics

Cardiac remodeling involves cellular and structural changes that occur as consequence of multifactorial events to maintain the homeostasis. The progression of pathological cardiac remodeling involves a transition from adaptive to maladaptive changes that eventually leads to impairment of ventricular function and heart failure. In this scenario, proteins are key elements that orchestrate molecular events as increased expression of fetal genes, neurohormonal and second messengers' activation, contractile dysfunction, rearrangement of the extracellular matrix and alterations in heart geometry. Mass spectrometry based-proteomics has emerged as a sound method to study protein dysregulation and identification of cardiac diseases biomarkers in plasma. In this review, we summarize the main findings related to large-scale proteome modulation of cardiac cells and extracellular matrix occurred during pathological cardiac remodeling. We describe the recent proteomic progresses in the selection of protein targets and introduce the renin-angiotensin system as an interesting target for the treatment of pathological cardiac remodeling.     

Acoustic response to a pheromonal cue in the arctiid moth Pyrrharctia Isabella

ABSTRACT. The acoustic properties of the clicks emitted in response to male courtship pheromone by female Pyrrharctia Isabella (J. E. Smith) (Lepidoptera: Arctiidae) have been investigated. Extracts of male scent organs or synthetic male pheromones can be applied to a glass rod and used to stimulate females to produce these sounds. Power spectra, sound pressure readings, and oscillographic analyses show that the acoustic signals elicited by male extracts or synthetic male pheromones are not distinguishable from those produced in response to disturbance (handling). This is the first reported example of a sound produced by a female moth in a sexual context, and is the first reported example in moths of an acoustic response to a pheromonal stimulus.