Novel Design of a Multimodal Technology-Based Smart Stethoscope for Personal Cardiovascular Health Monitoring.

Heart sounds and heart rate (pulse) are the most common physiological signals used in the diagnosis of cardiovascular diseases. Measuring these signals using a device and analyzing their interrelationships simultaneously can improve the accuracy of existing methods and propose new approaches for the diagnosis of cardiovascular diseases. In this study, we have presented a novel smart stethoscope based on multimodal physiological signal measurement technology for personal cardiovascular health monitoring. The proposed device is designed in the shape of a compact personal computer mouse for easy grasping and attachment to the surface of the chest using only one hand. A digital microphone and photoplehysmogram sensor are installed on the bottom and top surfaces of the device, respectively, to measure heart sound and pulse from the user's chest and finger simultaneously. In addition, a high-performance Bluetooth Low Energy System-on-Chip ARM microprocessor is used for pre-processing of measured data and communication with the smartphone. The prototype is assembled on a manufactured printed circuit board and 3D-printed shell to conduct an in vivo experiment to test the performance of physiological signal measurement and usability by observing users' muscle fatigue variation.

Automated Quality Assessment for Accelerometer-Based Heart Sounds Recorded with a Novel Subcutaneous Medical Implant.

In the context of monitoring patients with heart failure conditions, the automated assessment of heart sound quality is of major importance to insure the relevance of the medical analysis of the heart sound data. We propose in this study a technique of quality classification based on the selection of a small set of representative features. The first features are chosen to characterize whether the periodicity, complexity or statistical nature of the heart sound recordings. After segmentation process, the latter features are probing the detectability of the heart sounds in cardiac cycles. Our method is applied on a novel subcutaneous medical implant that combines ECG and accelerometric-based heart sound measurements. The actual prototype is in pre-clinical phase and has been implanted on 4 pigs, which anatomy and activity constitute a challenging environment for obtaining clean heart sounds. As reference quality labeling, we have performed a three-class manual annotation of each recording, qualified as "good", "unsure" and "bad". Our method allows to retrieve good quality heart sounds with a sensitivity and an accuracy of 82% ± 2% and 88% ± 6% respectively. Clinical Relevance- By accurately recovering high quality heart sound sequences, our method will enable to monitor reliable physiological indicators of heart failure complications such as decompensation.

Classification of heart sounds based on the combination of the modified frequency wavelet transform and convolutional neural network.

We purpose a novel method that combines modified frequency slice wavelet transform (MFSWT) and convolutional neural network (CNN) for classifying normal and abnormal heart sounds. A hidden Markov model is used to find the position of each cardiac cycle in the heart sound signal and determine the exact position of the four periods of S1, S2, systole, and diastole. Then the one-dimensional cardiac cycle signal was converted into a two-dimensional time-frequency picture using the MFSWT. Finally, two CNN models are trained using the aforementioned pictures. We combine two CNN models using sample entropy (SampEn) to determine which model is used to classify the heart sound signal. We evaluated our model on the heart sound public dataset provided by the PhysioNet Computing in Cardiology Challenge 2016. Experimental classification performance from a 10-fold cross-validation indicated that sensitivity (Se), specificity (Sp) and mean accuracy (MAcc) were 0.95, 0.93, and 0.94, respectively. The results showed the proposed method can classify normal and abnormal heart sounds with efficiency and high accuracy. Graphical abstract Block diagram of heart sound classification.

A Markov-Switching Model Approach to Heart Sound Segmentation and Classification.

OBJECTIVE: We consider challenges in accurate segmentation of heart sound signals recorded under noisy clinical environments for subsequent classification of pathological events. Existing state-of-the-art solutions to heart sound segmentation use probabilistic models such as hidden Markov models (HMMs), which, however, are limited by its observation independence assumption and rely on pre-extraction of noise-robust features. METHODS: We propose a Markov-switching autoregressive (MSAR) process to model the raw heart sound signals directly, which allows efficient segmentation of the cyclical heart sound states according to the distinct dependence structure in each state. To enhance robustness, we extend the MSAR model to a switching linear dynamic system (SLDS) that jointly model both the switching AR dynamics of underlying heart sound signals and the noise effects. We introduce a novel algorithm via fusion of switching Kalman filter and the duration-dependent Viterbi algorithm, which incorporates the duration of heart sound states to improve state decoding. RESULTS: Evaluated on Physionet/CinC Challenge 2016 dataset, the proposed MSAR-SLDS approach significantly outperforms the hidden semi-Markov model (HSMM) in heart sound segmentation based on raw signals and comparable to a feature-based HSMM. The segmented labels were then used to train Gaussian-mixture HMM classifier for identification of abnormal beats, achieving high average precision of 86.1% on the same dataset including very noisy recordings. CONCLUSION: The proposed approach shows noticeable performance in heart sound segmentation and classification on a large noisy dataset. SIGNIFICANCE: It is potentially useful in developing automated heart monitoring systems for pre-screening of heart pathologies.

An Automatic Approach Using ELM Classifier for HFpEF Identification Based on Heart Sound Characteristics.

Heart failure with preserved ejection fraction (HFpEF) is a complex and heterogeneous clinical syndrome. For the purpose of assisting HFpEF diagnosis, a non-invasive method using extreme learning machine and heart sound (HS) characteristics was provided in this paper. Firstly, the improved wavelet denoising method was used for signal preprocessing. Then, the logistic regression based hidden semi-Markov model algorithm was utilized to locate the boundary of the first HS and the second HS, therefore, the ratio of diastolic to systolic duration can be calculated. Eleven features were extracted based on multifractal detrended fluctuation analysis to analyze the differences of multifractal behavior of HS between healthy people and HFpEF patients. Afterwards, the statistical analysis was implemented on the extracted HS characteristics to generate the diagnostic feature set. Finally, the extreme learning machine was applied for HFpEF identification by the comparison of performances with support vector machine. The result shows an accuracy of 96.32%, a sensitivity of 95.48% and a specificity of 97.10%, which demonstrates the effectiveness of HS for HFpEF diagnosis.

Deep Convolutional Neural Networks for Heart Sound Segmentation.

This paper studies the use of deep convolutional neural networks to segment heart sounds into their main components. The proposed methods are based on the adoption of a deep convolutional neural network architecture, which is inspired by similar approaches used for image segmentation. Different temporal modeling schemes are applied to the output of the proposed neural network, which induce the output state sequence to be consistent with the natural sequence of states within a heart sound signal (S1, systole, S2, diastole). In particular, convolutional neural networks are used in conjunction with underlying hidden Markov models and hidden semi-Markov models to infer emission distributions. The proposed approaches are tested on heart sound signals from the publicly available PhysioNet dataset, and they are shown to outperform current state-of-the-art segmentation methods by achieving an average sensitivity of 93.9% and an average positive predictive value of 94% in detecting S1 and S2 sounds.

Adaptive Sojourn Time HSMM for Heart Sound Segmentation.

Heart sounds are difficult to interpret due to events with very short temporal onset between them (tens of milliseconds) and dominant frequencies that are out of the human audible spectrum. Computer-assisted decision systems may help but they require robust signal processing algorithms. In this paper, we propose a new algorithm for heart sound segmentation using a hidden semi-Markov model. The proposed algorithm infers more suitable sojourn time parameters than those currently suggested by the state of the art, through a maximum likelihood approach. We test our approach over three different datasets, including the publicly available PhysioNet and Pascal datasets. We also release a pediatric dataset composed of 29 heart sounds. In contrast with any other dataset available online, the annotations of the heart sounds in the released dataset contain information about the beginning and the ending of each heart sound event. Annotations were made by two cardiopulmonologists. The proposed algorithm is compared with the current state of the art. The results show a significant increase in segmentation performance, regardless the dataset or the methodology presented. For example, when using the PhysioNet dataset to train and to evaluate the HSMMs, our algorithm achieved average an F-score of [Formula: see text] compared to [Formula: see text] achieved by the algorithm described in [D.B. Springer, L. Tarassenko, and G. D. Clifford, "Logistic regressionHSMM-based heart sound segmentation," IEEE Transactions on Biomedical Engineering, vol. 63, no. 4, pp. 822-832, 2016]. In this sense, the proposed approach to adapt sojourn time parameters represents an effective solution for heart sound segmentation problems, even when the training data does not perfectly express the variability of the testing data.

Radar-Based Heart Sound Detection.

This paper introduces heart sound detection by radar systems, which enables touch-free and continuous monitoring of heart sounds. The proposed measurement principle entails two enhancements in modern vital sign monitoring. First, common touch-based auscultation with a phonocardiograph can be simplified by using biomedical radar systems. Second, detecting heart sounds offers a further feasibility in radar-based heartbeat monitoring. To analyse the performance of the proposed measurement principle, 9930 seconds of eleven persons-under-tests' vital signs were acquired and stored in a database using multiple, synchronised sensors: a continuous wave radar system, a phonocardiograph (PCG), an electrocardiograph (ECG), and a temperature-based respiration sensor. A hidden semi-Markov model is utilised to detect the heart sounds in the phonocardiograph and radar data and additionally, an advanced template matching (ATM) algorithm is used for state-of-the-art radar-based heartbeat detection. The feasibility of the proposed measurement principle is shown by a morphology analysis between the data acquired by radar and PCG for the dominant heart sounds S1 and S2: The correlation is 82.97 ± 11.15% for 5274 used occurrences of S1 and 80.72 ± 12.16% for 5277 used occurrences of S2. The performance of the proposed detection method is evaluated by comparing the F-scores for radar and PCG-based heart sound detection with ECG as reference: Achieving an F1 value of 92.22 ± 2.07%, the radar system approximates the score of 94.15 ± 1.61% for the PCG. The accuracy regarding the detection timing of heartbeat occurrences is analysed by means of the root-mean-square error: In comparison to the ATM algorithm (144.9 ms) and the PCG-based variant (59.4 ms), the proposed method has the lowest error value (44.2 ms). Based on these results, utilising the detected heart sounds considerably improves radar-based heartbeat monitoring, while the achieved performance is also competitive to phonocardiography.

Prognostic value of acoustic cardiography in patients with chronic heart failure.

BACKGROUND: Rapid risk stratification in patients with heart failure is critically important but challenging. The aim of our study is to ascertain whether acoustic cardiography can identify heart failure (HF) patients at high risk for mortality. METHODS: A total of 474 HF patients were enrolled into our study (76±11years old). Acoustic cardiographic parameters included S3 score (ie, third heart sound exists) and systolic dysfunction index (SDI) (correlated closely with left ventricular systolic dysfunction). The event-free survival curves were plotted by Kaplan-Meier method. Cox regression analysis was used to identify independent predictors for all-cause mortality. RESULTS: During a mean follow-up of 484days, 169 (35.7%) patients died and 126 (26.6%) were due to cardiac causes. After controlling for age, systolic blood pressure, hemoglobin, blood urea nitrogen, albumin, as well as ACEI and beta-blocker treatment in multivariate Cox regression analysis, SDI ≥5 and S3 score ≥4 were both independent predictors for all-cause mortality. Kaplan-Meier analysis showed that HF patients with SDI ≥5 or S3 score ≥4 had a significantly lower survival (52.2% vs. 69.2%, Log-rank χ(2)=18.07, P<0.001; 56.8% vs. 68.6%, Log-rank χ(2)=10.58, P=0.001, respectively) than those with lower SDI or S3 score. CONCLUSIONS: Acoustic cardiography could serve as a cost-effective and time-efficient tool to identify HF patients at high risk for mortality who might benefit from aggressive monitoring and intervention. It may improve assessment and initial disposition decisions in HF management.

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.

Why should you model time when you use Markov models for heart sound analysis.

Auscultation is a widely used technique in clinical activity to diagnose heart diseases. However, heart sounds are difficult to interpret because a) of events with very short temporal onset between them (tens of milliseconds) and b) dominant frequencies that are out of the human audible spectrum. In this paper, we propose a model to segment heart sounds using a semi-hidden Markov model instead of a hidden Markov model. Our model in difference from the state-of-the-art hidden Markov models takes in account the temporal constraints that exist in heart cycles. We experimentally confirm that semi-hidden Markov models are able to recreate the "true" continuous state sequence more accurately than hidden Markov models. We achieved a mean error rate per sample of 0.23.

Smartphone attachment for stethoscope recording.

With the ubiquity of smartphones and the rising technology of 3D printing, novel devices can be developed that leverage the "computer in your pocket" and rapid prototyping technologies toward scientific, medical, engineering, and creative purposes. This paper describes such a device: a simple 3D-printed extension for Apple's iPhone that allows the sound from an off-the-shelf acoustic stethoscope to be recorded using the phone's built-in microphone. The attachment's digital 3D files can be easily shared, modified for similar phones and devices capable of recording audio, and in combination with 3D printing technology allow for fabrication of a durable device without need for an entire factory of expensive and specialized machining tools. It is hoped that by releasing this device as an open source set of printable files that can be downloaded and reproduced cheaply, others can make use of these developments where access to cost-prohibitive, specialized medical instruments are not available. Coupled with specialized smartphone software ("apps"), more sophisticated and automated diagnostics may also be possible on-site.

Multi-center, multi-topic heart sound databases and their applications.

This paper describes a large resource of multi-center and multi-topic heart sound databases, which were based on the measured data from more than 9,000 heart sound samples (saved in WAV file format). According to different research topics, these samples were respectively stored in different folders (corresponding to different research topics and distributed over various cooperative research centers), most of which as subfolds were stored in a pooled folder in the principal center. According to different research topics, the measured data from these samples were used to create different databases. Relevant data for a specific topic can be pooled in a large database for further analysis. This resource is shared by members of related centers for their own specific topic. The applications of this resource include evaluation of cardiac safety of pregnant women, evaluation of cardiac reserve for children, athletes, addicts, astronauts, and general populations, as well as studies on a bedside method for evaluating cardiac energy, reversal of S1-S2 ratio, etc.

Phonocardiographic signal analysis method using a modified hidden Markov model.

Auscultation is an important diagnostic indicator for cardiovascular analysis. Heart sound classification and analysis play an important role in the auscultative diagnosis. This study uses a combination of Mel-frequency cepstral coefficient (MFCC) and hidden Markov model (HMM) to efficiently extract the features for pre-processed heart sound cycles for the purpose of classification. A system was developed for the interpretation of heart sounds acquired by phonocardiography using pattern recognition. The task of feature extraction was performed using three methods: time-domain feature, short-time Fourier transforms (STFT) and MFCC. The performances of these feature extraction methods were then compared. The results demonstrated that the proposed method using MFCC yielded improved interpretative information. Following the feature extraction, an automatic classification process was performed using HMM. Satisfactory classification results (sensitivity > or =0.952; specificity > or =0.953) were achieved for normal subjects and those with various murmur characteristics. These results were based on 1398 datasets obtained from 41 recruited subjects and downloaded from a public domain. Constituents characteristics of heart sounds were also evaluated using the proposed system. The findings herein suggest that the described system may have the potential to be used to assist doctors for a more objective diagnosis.

An assessment of the terminology used by diplomates and students to describe the character of equine mitral and aortic valve regurgitant murmurs: correlations with the physical properties of the sounds.

Twenty students and 16 diplomates listened to 7 recordings made from 7 horses with either aortic (n = 3) or mitral valve (n = 4) regurgitant murmurs. A total of 30 different terms were used to describe the character of these murmurs. However, only 4 terms were used in a repeatable and consistent manner. Most people described the character of a given mitral or aortic valve murmur with 1 or 2 terms. Diplomates drew from a pool of terms that was about half the size of that used by students--8.1 +/- 2.0 terms for diplomats (mean +/- 1 SD) versus 13.1 +/- 1.8 terms for students (P > .001). Only blowing, honking, buzzing, and musical were markedly associated with the recording played. Frequency analysis of the murmurs allowed them to be classified as containing harmonics (n = 4) or not containing harmonics (n = 3). Blowing was used to describe murmurs without harmonics on 39 of 48 occasions and corresponds to the term noisy used in some older descriptions of equine murmurs. Honking, musical, and buzzing were markedly associated with murmurs that contained harmonics; these terms were used 23, 13, and 12 of a possible 64 times, respectively. The frequency of buzzing and honking murmurs (72.7 +/- 9.3 and 88.4 +/- 46.3 Hz, respectively) was markedly lower than that of musical murmurs (156.8 +/- 81.1 Hz) (all P values < .01). Honking murmurs (0.392 +/- 0.092 seconds) were shorter than those described as buzzing or musical (0.496 +/- 0.205 and 0.504 +/- 0.116 seconds, respectively). The data suggest that the terminology for the character of aortic and mitral regurgitant murmurs should be restricted to 4 terms: blowing, honking, buzzing, and musical. Honking, buzzing, and musical describe murmurs with a peak dominant frequency and harmonics; blowing describes murmurs without a peak frequency. Effective communication could be enhanced by playing examples of reference sounds when these terms are taught so that nomenclature is used more uniformly.

Assessment of arterial distensibility by monitoring the timing of Korotkoff sounds.

The timing of Korotkoff sounds, blood pressure, and heart rate can now be monitored in the ambulatory patient: the QKD interval is the time between the onset of the depolarization on the electrocardiogram (Q) and detection of the last Korotkoff sound (K) at the level of brachial artery during cuff deflation, corresponding to diastolic blood pressure (D). Because this interval is inversely related to pulse wave velocity, this recently developed device enables evaluation of the influence of blood pressure on arterial rigidity, providing valuable information on the properties of the arteries. In this study, we examined the influence of hypertension and age on the above parameters and their correlations to left ventricular mass. QKD interval, blood pressure, and heart rate were monitored over a period of 24 h (four measurements/hour) in 33 normotensive and 70 untreated essential hypertensive patients. The slopes of the plots of QKD interval versus systolic and pulse pressure during the 24 h were calculated for each patient. The influence of age and hypertension on these slopes was tested by comparison of matched groups and multivariate analysis. Moreover the relationships between these parameters and echocardiographically assessed left ventricular mass were studied in 37 patients. We found a reduction in mean QKD interval with age and hypertension, reflecting the recognized higher pulse wave velocity in these patients. The slopes of the plots of QKD interval versus blood pressure were also lower in these patients, indicating the smaller influence of a change in blood pressure on pulse wave velocity in patients with stiffer arteries.(ABSTRACT TRUNCATED AT 250 WORDS)