-
Annual International Conference of the... Nov 2021Cardiac auscultation is the key screening procedure to detect and identify cardiovascular diseases (CVDs). One of many steps to automatically detect CVDs using...
Cardiac auscultation is the key screening procedure to detect and identify cardiovascular diseases (CVDs). One of many steps to automatically detect CVDs using auscultation, concerns the detection and delimitation of the heart sound boundaries, a process known as segmentation. Whether to include or not a segmentation step in the signal classification pipeline is nowadays a topic of discussion. Up to our knowledge, the outcome of a segmentation algorithm has been used almost exclusively to align the different signal segments according to the heartbeat. In this paper, the need for a heartbeat alignment step is tested and evaluated over different machine learning algorithms, including deep learning solutions. From the different classifiers tested, Gate Recurrent Unit (GRU) Network and Convolutional Neural Network (CNN) algorithms are shown to be the most robust. Namely, these algorithms can detect the presence of heart murmurs even without a heartbeat alignment step. Furthermore, Support Vector Machine (SVM) and Random Forest (RF) algorithms require an explicit segmentation step to effectively detect heart sounds and murmurs, the overall performance is expected drop approximately 5% on both cases.
Topics: Algorithms; Heart Auscultation; Heart Sounds; Neural Networks, Computer; Support Vector Machine
PubMed: 34891292
DOI: 10.1109/EMBC46164.2021.9630559 -
Methods (San Diego, Calif.) Jun 2022This paper presents a heart murmur detection and multi-class classification approach via machine learning. We extracted heart sound and murmur features that are of...
This paper presents a heart murmur detection and multi-class classification approach via machine learning. We extracted heart sound and murmur features that are of diagnostic importance and developed additional 16 features that are not perceivable by human ears but are valuable to improve murmur classification accuracy. We examined and compared the classification performance of supervised machine learning with k-nearest neighbor (KNN) and support vector machine (SVM) algorithms. We put together a test repertoire having more than 450 heart sound and murmur episodes to evaluate the performance of murmur classification using cross-validation of 80-20 and 90-10 splits. As clearly demonstrated in our evaluation, the specific set of features chosen in our study resulted in accurate classification consistently exceeding 90% for both classifiers.
Topics: Algorithms; Heart Murmurs; Heart Sounds; Humans; Machine Learning; Support Vector Machine
PubMed: 34245871
DOI: 10.1016/j.ymeth.2021.07.002 -
IEEE Transactions on Biomedical... Dec 2019Heart-sound auscultation is a rapid and fundamental technique used for examining the cardiovascular system. The main components of heart sounds are the first and second...
Heart-sound auscultation is a rapid and fundamental technique used for examining the cardiovascular system. The main components of heart sounds are the first and second heart sounds. Discriminating these heart sounds under the presence of additional heart sounds and murmurs will be difficult. To recognize these signals efficiently, this study proposes a monitoring system with phonocardiogram and electrocardiogram. This system has two key points. The first is chip implementation, including capacitor coupled amplifier, transimpedance amplifier, high-pass sigma-delta modulator, and digital signal processing block. The chip in the system is fabricated in 0.18 μm standard complementary metal-oxide-semiconductor process. The second is a software application on smartphones for heart-related physiological signal recording, display, and identification. A wavelet-based QRS complex detection algorithm verified by MIT/BIH Arrhythmia Database is also proposed. The overall measured positive prediction, sensitivity, and error rate of the proposed algorithm are 99.90%, 99.82%, and 0.28%, respectively. During auscultation, doctors may refer to these physiological signals displayed on the smartphone and simultaneously listen to the heart sounds to diagnose the potential heart disease. By taking advantage of signal visualization and keeping the original diagnosis procedure, the uncertainty existing in heart sounds can be eliminated, and the training period to acquire auscultation skills can be reduced.
Topics: Algorithms; Amplifiers, Electronic; Cardiovascular Diseases; Electrocardiography; Heart Auscultation; Humans; Phonocardiography; Semiconductors; Signal Processing, Computer-Assisted; Wearable Electronic Devices
PubMed: 31634841
DOI: 10.1109/TBCAS.2019.2947694 -
Frontiers in Pediatrics 2022Still's murmur is the most prevalent innocent heart murmur of childhood. Auscultation is the primary clinical tool to identify this murmur as innocent. Whereas pediatric...
BACKGROUND
Still's murmur is the most prevalent innocent heart murmur of childhood. Auscultation is the primary clinical tool to identify this murmur as innocent. Whereas pediatric cardiologists routinely perform this task, primary care providers are less successful in distinguishing Still's murmur from the murmurs of true heart disease. This results in a large number of children with a Still's murmur being referred to pediatric cardiologists.
OBJECTIVES
To develop a computer algorithm that can aid primary care providers to identify the innocent Still's murmur at the point of care, to substantially decrease over-referral.
METHODS
The study included Still's murmurs, pathological murmurs, other innocent murmurs, and normal (i.e., non-murmur) heart sounds of 1,473 pediatric patients recorded using a commercial electronic stethoscope. The recordings with accompanying clinical diagnoses provided by a pediatric cardiologist were used to train and test the convolutional neural network-based algorithm.
RESULTS
A comparative analysis showed that the algorithm using only the murmur sounds recorded at the lower left sternal border achieved the highest accuracy. The developed algorithm identified Still's murmur with 90.0% sensitivity and 98.3% specificity for the default decision threshold. The area under the receiver operating characteristic curve was 0.943.
CONCLUSIONS
Still's murmur can be identified with high accuracy with the algorithm we developed. Using this approach, the algorithm could help to reduce the rate of unnecessary pediatric cardiologist referrals and use of echocardiography for a common benign finding.
PubMed: 36210944
DOI: 10.3389/fped.2022.923956 -
IEEE Journal of Biomedical and Health... Jun 2020This paper proposes a novel framework for the segmentation of phonocardiogram (PCG) signals into heart states, exploiting the temporal evolution of the PCG as well as...
OBJECTIVE
This paper proposes a novel framework for the segmentation of phonocardiogram (PCG) signals into heart states, exploiting the temporal evolution of the PCG as well as considering the salient information that it provides for the detection of the heart state.
METHODS
We propose the use of recurrent neural networks and exploit recent advancements in attention based learning to segment the PCG signal. This allows the network to identify the most salient aspects of the signal and disregard uninformative information.
RESULTS
The proposed method attains state-of-the-art performance on multiple benchmarks including both human and animal heart recordings. Furthermore, we empirically analyse different feature combinations including envelop features, wavelet and Mel Frequency Cepstral Coefficients (MFCC), and provide quantitative measurements that explore the importance of different features in the proposed approach.
CONCLUSION
We demonstrate that a recurrent neural network coupled with attention mechanisms can effectively learn from irregular and noisy PCG recordings. Our analysis of different feature combinations shows that MFCC features and their derivatives offer the best performance compared to classical wavelet and envelop features.
SIGNIFICANCE
Heart sound segmentation is a crucial pre-processing step for many diagnostic applications. The proposed method provides a cost effective alternative to labour extensive manual segmentation, and provides a more accurate segmentation than existing methods. As such, it can improve the performance of further analysis including the detection of murmurs and ejection clicks. The proposed method is also applicable for detection and segmentation of other one dimensional biomedical signals.
Topics: Animals; Deep Learning; Female; Heart Sounds; Humans; Male; Neural Networks, Computer; Phonocardiography; Signal Processing, Computer-Assisted
PubMed: 31670683
DOI: 10.1109/JBHI.2019.2949516 -
Acta Veterinaria Scandinavica Jun 2020The frequency that cardiac murmurs are identified and recorded in first opinion veterinary practices at the first health check in puppies is unknown. The aims of the...
BACKGROUND
The frequency that cardiac murmurs are identified and recorded in first opinion veterinary practices at the first health check in puppies is unknown. The aims of the study were to assess the agreement between first opinion veterinary practitioners, a veterinary student and a veterinary cardiology specialist on detecting murmurs, and to establish whether abnormal auscultation findings had been recorded in the health certificates of clinically healthy puppies. The study included prospective and retrospective investigations, where the prospectively collected auscultation findings from a veterinary cardiology specialist and a trained veterinary student were compared to auscultation findings recorded by first opinion veterinary practitioners.
RESULTS
Cardiac auscultation was performed on 331 client-owned, clinically healthy dogs at two time points: at age 34-69 days by a first opinion veterinary practitioner and at age 45-76 days, on average 9 days later, by a veterinary cardiology specialist and a trained veterinary student. Agreement among the three was compared for the presence of a murmur. The degree of inter-observer agreement was evaluated using Cohen's kappa. Auscultation findings, as noted in the pets' passports, from 331 puppies and 43 different first opinion veterinary practices, were retrospectively reviewed and prospectively compared with auscultation findings from a veterinary cardiology specialist. Agreement between the veterinary cardiology specialist and the first opinion veterinary practitioners was poor (ϰ = 0.01) and significantly different (P < 0.001). First opinion veterinary practitioners had recorded a cardiac murmur in only 1 of the 97 puppies in which the veterinary cardiology specialist detected a murmur. Two-hundred-and-fifty-two puppies were auscultated by both the veterinary cardiology specialist and the student. Their agreement was fair (ϰ = 0.40) and significantly different (P = 0.024). The agreement between the student and a first opinion veterinary practitioner on these 252 puppies was poor (ϰ = 0.03) and significantly different (P < 0.001).
CONCLUSIONS
This study shows that soft cardiac murmurs are rarely documented during the first veterinary health check in puppies by first opinion veterinary practitioners. Although soft murmurs may not be clinically relevant, finding and recording them is evidence of a carefully performed auscultation. Missing a non-pathological murmur is not of clinical importance; however, missing a pathological murmur could prove detrimental for the individual puppy.
Topics: Animals; Dog Diseases; Dogs; Heart Auscultation; Heart Murmurs; Prospective Studies; Retrospective Studies; Students; Veterinarians; Veterinary Medicine
PubMed: 32586343
DOI: 10.1186/s13028-020-00535-1 -
Journal of Veterinary Cardiology : the... Jun 2022A three-month-old, male intact Norwegian forest cat without any clinical signs was referred to the cardiology service of the author's teaching hospital for evaluation of...
A three-month-old, male intact Norwegian forest cat without any clinical signs was referred to the cardiology service of the author's teaching hospital for evaluation of a cardiac murmur. The murmur was systolic with an intensity of 4 out of 6 with the point of maximal intensity at the left heart base. Echocardiography revealed a moderate mitral valve regurgitation and a moderate dynamic left ventricular outflow tract obstruction both resulting from systolic anterior motion of the mitral valve (SAM). Moreover, left ventricular concentric hypertrophy was noted. Oral atenolol therapy was initiated. Recheck examination 3.5 months later revealed unchanged murmur characteristics in the still asymptomatic kitten. Echocardiography showed no SAM, but there was a severe fixed aortic stenosis apparent caused by a discrete supravalvular lesion, 4 mm distal to the valve, with an hourglass morphology. Supravalvular aortic stenosis is a rare congenital anomaly in cats, which has not been reported antemortem yet.
Topics: Animals; Aortic Stenosis, Supravalvular; Cat Diseases; Cats; Echocardiography; Female; Heart Murmurs; Hypertrophy, Left Ventricular; Male; Mitral Valve; Mitral Valve Insufficiency
PubMed: 35567886
DOI: 10.1016/j.jvc.2022.04.001 -
Animals : An Open Access Journal From... Apr 2024Auscultation of heart sounds is an important veterinary skill requiring an understanding of anatomy, physiology, pathophysiology and pattern recognition. This...
Auscultation of heart sounds is an important veterinary skill requiring an understanding of anatomy, physiology, pathophysiology and pattern recognition. This cross-sectional study was developed to evaluate a targeted, audio-visual training resource for veterinary students to improve their understanding and auscultation of common heart conditions in horses. Fourth- and fifth-year 2021 and 2022 Bachelor of Veterinary Science students at the University of Queensland (UQ) were provided the learning resource and surveyed via online pre- and post-intervention surveys. Results were quantitatively analyzed using descriptive statistics and Mann-Whitney U tests. Open-ended survey questions were qualitatively analyzed by thematic analysis and Leximancer™ Version 4 program software analysis. Over the two-year period, 231 fourth-year and 222 fifth-year veterinary students had access to the resource; 89 completed the pre-intervention survey and 57 completed the post-intervention survey. Quantitative results showed the resource helped students prepare for practicals and their perception of competency and confidence when auscultating equine cardiac sounds improved ( < 0.05). Compared to fifth-year students, fourth-year students felt less competent at identifying murmurs and arrythmias prior to accessing the learning resource ( < 0.05). Fourth-year and fifth-year students' familiarity with detection of murmurs improved after completing the learning resource ( < 0.001). Qualitative analysis demonstrated a limited number of opportunities to practice equine cardiac auscultation throughout the veterinary degree, especially during the COVID-19 pandemic, and that integrated audio-visual resources are an effective means of teaching auscultation.
PubMed: 38731348
DOI: 10.3390/ani14091341 -
PLOS Digital Health Mar 2024Cardiovascular diseases (CVDs) account for a high fatality rate worldwide. Heart murmurs can be detected from phonocardiograms (PCGs) and may indicate CVDs. Still, they...
OBJECTIVE
Cardiovascular diseases (CVDs) account for a high fatality rate worldwide. Heart murmurs can be detected from phonocardiograms (PCGs) and may indicate CVDs. Still, they are often overlooked as their detection and correct clinical interpretation require expert skills. In this work, we aim to predict the presence of murmurs and clinical outcomes from multiple PCG recordings employing an explainable multitask model.
APPROACH
Our approach consists of a two-stage multitask model. In the first stage, we predict the murmur presence in single PCGs using a multiple instance learning (MIL) framework. MIL also allows us to derive sample-wise classifications (i.e. murmur locations) while only needing one annotation per recording ("weak label") during training. In the second stage, we fuse explainable hand-crafted features with features from a pooling-based artificial neural network (PANN) derived from the MIL framework. Finally, we predict the presence of murmurs and the clinical outcome for a single patient based on multiple recordings using a simple feed-forward neural network.
MAIN RESULTS
We show qualitatively and quantitatively that the MIL approach yields useful features and can be used to detect murmurs on multiple time instances and may thus guide a practitioner through PCGs. We analyze the second stage of the model in terms of murmur classification and clinical outcome. We achieved a weighted accuracy of 0.714 and an outcome cost of 13612 when using the PANN model and demographic features on the CirCor dataset (hidden test set of the George B. Moody PhysioNet challenge 2022, team "Heart2Beat", rank 12 / 40).
SIGNIFICANCE
To the best of our knowledge, we are the first to demonstrate the usefulness of MIL in PCG classification. Also, we showcase how the explainability of the model can be analyzed quantitatively, thus avoiding confirmation bias inherent to many post-hoc methods. Finally, our overall results demonstrate the merit of employing MIL combined with handcrafted features for the generation of explainable features as well as for a competitive classification performance.
PubMed: 38502666
DOI: 10.1371/journal.pdig.0000461 -
IEEE Journal of Biomedical and Health... Sep 2023Cardiac auscultation, exhibited by phonocardiogram (PCG), is a non-invasive and low-cost diagnostic method for cardiovascular diseases (CVDs). However, deploying it in...
Cardiac auscultation, exhibited by phonocardiogram (PCG), is a non-invasive and low-cost diagnostic method for cardiovascular diseases (CVDs). However, deploying it in practice is quite challenging, due to the inherent murmurs and a limited number of supervised samples in heart sound data. To solve these problems, not only heart sound analysis based on handcrafted features, but also computer-aided heart sound analysis based on deep learning have been extensively studied in recent years. Though with elaborate design, most of these methods still use additional pre-processing to improve classification performance, which heavily relies on time-consuming experienced engineering. In this article, we propose a parameter-efficient densely connected dual attention network (DDA) for heart sound classification. It combines two advantages simultaneously of the purely end-to-end architecture and enriched contextual representations of the self-attention mechanism. Specifically, the densely connected structure can automatically extract the information flow of heart sound features hierarchically. Alongside, improving contextual modeling capabilities, the dual attention mechanism adaptively aggregates local features with global dependencies via a self-attention mechanism, which captures the semantic interdependencies across position and channel axes respectively. Extensive experiments across stratified 10-fold cross-validation strongly evidence that our proposed DDA model surpasses current 1D deep models on the challenging Cinc2016 benchmark with significant computational efficiency.
Topics: Humans; Heart Murmurs; Heart Sounds; Heart Auscultation; Cardiovascular Diseases
PubMed: 37318972
DOI: 10.1109/JBHI.2023.3286585