The CirCor DigiScope Dataset: From Murmur Detection to Murmur Classification

Rapid detection and interpretation of heart murmurs using phonocardiograms, transfer learning and explainable artificial intelligence

Cardiovascular disease, which remains one of the main causes of death, can be prevented by early diagnosis of heart sounds. Certain noisy signals, known as murmurs, may be present in heart sounds. On auscultation, the degree of murmur is closely related to the patient's clinical condition. Computer-aided decision-making systems can help doctors to detect murmurs and make faster decisions. The Mel spectrograms were generated from raw phonocardiograms and then presented to the OpenL3 network for transfer learning. In this way, the signals were classified to predict the presence or absence of murmurs and their level of severity. Pitch level (healthy, low, medium, high) and Levine scale (healthy, soft, loud) were used. The results obtained without prior segmentation are very impressive. The model used was then interpreted using an Explainable Artificial Intelligence (XAI) method, Occlusion Sensitivity. This approach shows that XAI methods are necessary to know the features used internally by the artificial neural network then to explain the automatic decision taken by the model. The averaged image of the occlusion sensitivity maps can give us either an overview or a precise detail per pixel of the features used. In the field of healthcare, particularly cardiology, for rapid diagnostic and preventive purposes, this work could provide more detail on the important features of the phonocardiogram.

Cardiac murmur grading and risk analysis of cardiac diseases based on adaptable heterogeneous-modality multi-task learning

Cardiovascular disease (CVDs) has become one of the leading causes of death, posing a significant threat to human life. The development of reliable Artificial Intelligence (AI) assisted diagnosis algorithms for cardiac sounds is of great significance for early detection and treatment of CVDs. However, there is scarce research in this field. Existing research mainly faces three major challenges: (1) They mainly limited to murmur classification and cannot achieve murmur grading, but attempting both classification and grading may lead to negative effects between different multi-tasks. (2) They mostly pay attention to unstructured cardiac sound modality and do not consider the structured demographic modality, as it is difficult to balance the influence of heterogeneous modalities. (3) Deep learning methods lack interpretability, which makes it challenging to apply them clinically. To tackle these challenges, we propose a method for cardiac murmur grading and cardiac risk analysis based on heterogeneous modality adaptive multi-task learning. Specifically, a Hierarchical Multi-Task learning-based cardiac murmur detection and grading method (HMT) is proposed to prevent negative interference between different tasks. In addition, a cardiac risk analysis method based on Heterogeneous Multi-modal feature impact Adaptation (HMA) is also proposed, which transforms unstructured modality into structured modality representation, and utilizes an adaptive mode weight learning mechanism to balance the impact between unstructured modality and structured modality, thus enhancing the performance of cardiac risk prediction. Finally, we propose a multi-task interpretability learning module that incorporates an important evaluation using random masks. This module utilizes SHAP graphs to visualize crucial murmur segments in cardiac sound and employs a multi-factor risk decoupling model based on nomograms. And then we gain insights into the cardiac disease risk in both pre-decoupled multi-modality and post-decoupled single-modality scenarios, thus providing a solid foundation for AI assisted cardiac murmur grading and risk analysis. Experimental results on a large real-world CirCor DigiScope PCG dataset demonstrate that the proposed method outperforms the state-of-the-art (SOTA) method in murmur detection, grading, and cardiac risk analysis, while also providing valuable diagnostic evidence.

External evaluation of a commercial artificial intelligence-augmented digital auscultation platform in valvular heart disease detection using echocardiography as reference standard

OBJECTIVE

There are few studies evaluating the accuracy of commercially available AI-powered digital auscultation platforms in detecting valvular heart disease (VHD). Therefore, the utility of these systems for diagnosing clinically significant VHD remains unclear. We conducted a comprehensive external evaluation of the Eko murmur analysis software (EMAS) and report its accuracy in detecting murmurs associated with VHD using echocardiography (ECHO) as the reference standard.

METHODS

We analyzed phonocardiogram (PCG) and ECHO data from 1,029 individuals (461 females, mean (SD) age: 61 (29) years, BMI: 29 (9)) at a single academic medical center. PCGs were recorded using the EkoDUO and EkoCORE stethoscopes from the four standard auscultation positions immediately before transthoracic ECHO (TTE) testing. TTE diagnostics were used as reference to calculate the EMAS sensitivity and specificity in detecting murmurs associated with VHD. The 95% confidence intervals are reported.

RESULTS

Of the 4,081 PCGs, 79% were of sufficient quality for murmur analysis. The sensitivity and specificity of the EMAS in detecting VHD were 39.3% (95% CI: 37.2-41.3) and 82.3% (95% CI: 80.0-84.5), respectively. EMAS sensitivity in detecting murmurs associated with common VHD types was 62.5%, 75.0%, 88.9%, and 63.3% for moderate-severe and severe cases of mitral stenosis, aortic regurgitation, aortic stenosis, and mitral regurgitation, respectively.

CONCLUSION

The EMAS algorithm exhibits limited overall sensitivity in detecting VHD. The sensitivity of the algorithm varies across VHD types. These findings suggest that EMAS can be used for diagnosis of specific lesions, but not all VHD types, which limits its clinical applicability as a screening tool.

Phonocardiogram (PCG) Murmur Detection Based on the Mean Teacher Method

Cardiovascular diseases (CVDs) are among the primary causes of mortality globally, highlighting the critical need for early detection to mitigate their impact. Phonocardiograms (PCGs), which record heart sounds, are essential for the non-invasive assessment of cardiac function, enabling the early identification of abnormalities such as murmurs. Particularly in underprivileged regions with high birth rates, the absence of early diagnosis poses a significant public health challenge. In pediatric populations, the analysis of PCG signals is invaluable for detecting abnormal sound waves indicative of congenital and acquired heart diseases, such as septal defects and defective cardiac valves. In the PhysioNet 2022 challenge, the murmur score is a weighted accuracy metric that reflects detection accuracy based on clinical significance. In our research, we proposed a mean teacher method tailored for murmur detection, making full use of the Phyionet2022 and Phyionet2016 PCG datasets, achieving the SOTA (State of Art) performance with a murmur score of 0.82 and an AUC score of 0.90, providing an accessible and high accuracy non-invasive early stage CVD assessment tool, especially for low and middle-income countries (LMICs).

Deep Learning in Heart Sound Analysis: From Techniques to Clinical Applications

Importance: Heart sound auscultation is a routinely used physical examination in clinical practice to identify potential cardiac abnormalities. However, accurate interpretation of heart sounds requires specialized training and experience, which limits its generalizability. Deep learning, a subset of machine learning, involves training artificial neural networks to learn from large datasets and perform complex tasks with intricate patterns. Over the past decade, deep learning has been successfully applied to heart sound analysis, achieving remarkable results and accumulating substantial heart sound data for model training. Although several reviews have summarized deep learning algorithms for heart sound analysis, there is a lack of comprehensive summaries regarding the available heart sound data and the clinical applications. Highlights: This review will compile the commonly used heart sound datasets, introduce the fundamentals and state-of-the-art techniques in heart sound analysis and deep learning, and summarize the current applications of deep learning for heart sound analysis, along with their limitations and areas for future improvement. Conclusions: The integration of deep learning into heart sound analysis represents a significant advancement in clinical practice. The growing availability of heart sound datasets and the continuous development of deep learning techniques contribute to the improvement and broader clinical adoption of these models. However, ongoing research is needed to address existing challenges and refine these technologies for broader clinical use.

Improved recovery of cardiac auscultation sounds using modified cosine transform and LSTM-based masking

Obtaining accurate cardiac auscultation signals, including basic heart sounds (S1 and S2) and subtle signs of disease, is crucial for improving cardiac diagnoses and making the most of telehealth. This research paper introduces an innovative approach that utilizes a modified cosine transform (MCT) and a masking strategy based on long short-term memory (LSTM) to effectively distinguish heart sounds and murmurs from background noise and interfering sounds. The MCT is used to capture the repeated pattern of the heart sounds, while the LSTMs are trained to construct masking based on the repeated MCT spectrum. The proposed strategy's performance in maintaining the clinical relevance of heart sounds continues to demonstrate effectiveness, even in environments marked by increased noise and complex disruptions. The present work highlights the clinical significance and reliability of the suggested methodology through in-depth signal visualization and rigorous statistical performance evaluations. In comparative assessments, the proposed approach has demonstrated superior performance compared to recent algorithms, such as LU-Net and PC-DAE. Furthermore, the system's adaptability to various datasets enhances its reliability and practicality. The suggested method is a potential way to improve the accuracy of cardiovascular diagnostics in an era of rapid advancement in medical signal processing. The proposed approach showed an enhancement in the average signal-to-noise ratio (SNR) by 9.6 dB at an input SNR of - 6 dB and by 3.3 dB at an input SNR of 10 dB. The average signal distortion ratio (SDR) achieved across a variety of input SNR values was 8.56 dB.

Identifying pediatric heart murmurs and distinguishing innocent from pathologic using deep learning

OBJECTIVE

To develop a deep learning algorithm to perform multi-class classification of normal pediatric heart sounds, innocent murmurs, and pathologic murmurs.

METHODS

We prospectively enrolled children under age 18 being evaluated by the Division of Pediatric Cardiology. Parents provided consent for a deidentified recording of their child's heart sounds with a digital stethoscope. Innocent murmurs were validated by a pediatric cardiologist and pathologic murmurs were validated by echocardiogram. To augment our collection of normal heart sounds, we utilized a public database of pediatric heart sound recordings (Oliveira, 2022). We propose two novel approaches for this audio classification task. We train a vision transformer on either Markov transition field or Gramian angular field image representations of the frequency spectrum. We benchmark our results against a ResNet-50 CNN trained on spectrogram images.

RESULTS

Our final dataset consisted of 366 normal heart sounds, 175 innocent murmurs, and 216 pathologic murmurs. Innocent murmurs collected include Still's murmur, venous hum, and flow murmurs. Pathologic murmurs included ventricular septal defect, tetralogy of Fallot, aortic regurgitation, aortic stenosis, pulmonary stenosis, mitral regurgitation and stenosis, and tricuspid regurgitation. We find that the Vision Transformer consistently outperforms the ResNet-50 on all three image representations, and that the Gramian angular field is the superior image representation for pediatric heart sounds. We calculated a one-vs-rest multi-class ROC curve for each of the three classes. Our best model achieves an area under the curve (AUC) value of 0.92 ± 0.05, 0.83 ± 0.04, and 0.88 ± 0.04 for identifying normal heart sounds, innocent murmurs, and pathologic murmurs, respectively.

CONCLUSION

We present two novel methods for pediatric heart sound classification, which outperforms the current standard of using a convolutional neural network trained on spectrogram images. To our knowledge, we are the first to demonstrate multi-class classification of pediatric murmurs. Multiclass output affords a more explainable and interpretable model, which can facilitate further model improvement in the downstream model development cycle and enhance clinician trust and therefore adoption.

Recognition of Patient Gender: A Machine Learning Preliminary Analysis Using Heart Sounds from Children and Adolescents

Research has shown that X-rays and fundus images can classify gender, age group, and race, raising concerns about bias and fairness in medical AI applications. However, the potential for physiological sounds to classify sociodemographic traits has not been investigated. Exploring this gap is crucial for understanding the implications and ensuring fairness in the field of medical sound analysis. We aimed to develop classifiers to determine gender (men/women) based on heart sound recordings and using machine learning (ML). Data-driven ML analysis. We utilized the open-access CirCor DigiScope Phonocardiogram Dataset obtained from cardiac screening programs in Brazil. Volunteers < 21 years of age. Each participant completed a questionnaire and underwent a clinical examination, including electronic auscultation at four cardiac points: aortic (AV), mitral (MV), pulmonary (PV), and tricuspid (TV). We used Mel-frequency cepstral coefficients (MFCCs) to develop the ML classifiers. From each patient and from each auscultation sound recording, we extracted 10 MFCCs. In sensitivity analysis, we additionally extracted 20, 30, 40, and 50 MFCCs. The most effective gender classifier was developed using PV recordings (AUC ROC = 70.3%). The second best came from MV recordings (AUC ROC = 58.8%). AV and TV recordings produced classifiers with an AUC ROC of 56.4% and 56.1%, respectively. Using more MFCCs did not substantially improve the classifiers. It is possible to classify between males and females using phonocardiogram data. As health-related audio recordings become more prominent in ML applications, research is required to explore if these recordings contain signals that could distinguish sociodemographic features.

An open-access simultaneous electrocardiogram and phonocardiogram database

Objective.The EPHNOGRAM project aimed to develop a low-cost, low-power device for simultaneous electrocardiogram (ECG) and phonocardiogram (PCG) recording, with additional channels for environmental audio to enhance PCG through active noise cancellation. The objective was to study multimodal electro-mechanical activities of the heart, offering insights into the differences and synergies between these modalities during various cardiac activity levels.Approach.We developed and tested several hardware prototypes of a simultaneous ECG-PCG acquisition device. Using this technology, we collected simultaneous ECG and PCG data from 24 healthy adults during different physical activities, including resting, walking, running, and stationary biking, in an indoor fitness center. The data were annotated using a robust software that we developed for detecting ECG R-peaks and PCG S1 and S2 components, and overseen by a human expert. We also developed machine learning models using ECG-based, PCG-based, and joint ECG-PCG features, like R-R and S1-S2 intervals, to classify physical activities and analyze electro-mechanical dynamics.Main results.The results show a significant coupling between ECG and PCG components, especially during high-intensity exercise. Notable micro-variations in S2-based heart rate show differences in the heart's electrical and mechanical functions. The Lomb-Scargle periodogram and approximate entropy analyses confirm the higher volatility of S2-based heart rate compared to ECG-based heart rate. Correlation analysis shows stronger coupling between R-R and R-S1 intervals during high-intensity activities. Hybrid ECG-PCG features, like the R-S2 interval, were identified as more informative for physical activity classification through mRMR feature selection and SHAP value analysis.Significance.The EPHNOGRAM database, is available on PhysioNet. The database enhances our understanding of cardiac function, enabling future studies on the heart's mechanical and electrical interrelationships. The results of this study can contribute to improved cardiac condition diagnoses. Additionally, the designed hardware has the potential for integration into wearable devices and the development of multimodal stress test technologies.

Identification of Congenital Valvular Murmurs in Young Patients Using Deep Learning-Based Attention Transformers and Phonocardiograms

One in every four newborns suffers from congenital heart disease (CHD) that causes defects in the heart structure. The current gold-standard assessment technique, echocardiography, causes delays in the diagnosis owing to the need for experts who vary markedly in their ability to detect and interpret pathological patterns. Moreover, echo is still causing cost difficulties for low- and middle-income countries. Here, we developed a deep learning-based attention transformer model to automate the detection of heart murmurs caused by CHD at an early stage of life using cost-effective and widely available phonocardiography (PCG). PCG recordings were obtained from 942 young patients at four major auscultation locations, including the aortic valve (AV), mitral valve (MV), pulmonary valve (PV), and tricuspid valve (TV), and they were annotated by experts as absent, present, or unknown murmurs. A transformation to wavelet features was performed to reduce the dimensionality before the deep learning stage for inferring the medical condition. The performance was validated through 10-fold cross-validation and yielded an average accuracy and sensitivity of 90.23 % and 72.41 %, respectively. The accuracy of discriminating between murmurs' absence and presence reached 76.10 % when evaluated on unseen data. The model had accuracies of 70 %, 88 %, and 86 % in predicting murmur presence in infants, children, and adolescents, respectively. The interpretation of the model revealed proper discrimination between the learned attributes, and AV channel was found important (score 0.75) for the murmur absence predictions while MV and TV were more important for murmur presence predictions. The findings potentiate deep learning as a powerful front-line tool for inferring CHD status in PCG recordings leveraging early detection of heart anomalies in young people. It is suggested as a tool that can be used independently from high-cost machinery or expert assessment.

Murmur identification and outcome prediction in phonocardiograms using deep features based on Stockwell transform

Traditionally, heart murmurs are diagnosed through cardiac auscultation, which requires specialized training and experience. The purpose of this study is to predict patients' clinical outcomes (normal or abnormal) and identify the presence or absence of heart murmurs using phonocardiograms (PCGs) obtained at different auscultation points. A semi-supervised model tailored to PCG classification is introduced in this study, with the goal of improving performance using time-frequency deep features. The study begins by investigating the behavior of PCGs in the time-frequency domain, utilizing the Stockwell transform to convert the PCG signal into two-dimensional time-frequency maps (TFMs). A deep network named AlexNet is then used to derive deep feature sets from these TFMs. In feature reduction, redundancy is eliminated and the number of deep features is reduced to streamline the feature set. The effectiveness of the extracted features is evaluated using three different classifiers using the CinC/Physionet challenge 2022 dataset. For Task I, which focuses on heart murmur detection, the proposed approach achieved an average accuracy of 93%, sensitivity of 91%, and F1-score of 91%. According to Task II of the CinC/Physionet challenge 2022, the approach showed a clinical outcome cost of 5290, exceeding the benchmark set by leading methods in the challenge.

Multiple instance learning framework can facilitate explainability in murmur detection

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.

Exploring new horizons in neuroscience disease detection through innovative visual signal analysis

Brain disorders pose a substantial global health challenge, persisting as a leading cause of mortality worldwide. Electroencephalogram (EEG) analysis is crucial for diagnosing brain disorders, but it can be challenging for medical practitioners to interpret complex EEG signals and make accurate diagnoses. To address this, our study focuses on visualizing complex EEG signals in a format easily understandable by medical professionals and deep learning algorithms. We propose a novel time-frequency (TF) transform called the Forward-Backward Fourier transform (FBFT) and utilize convolutional neural networks (CNNs) to extract meaningful features from TF images and classify brain disorders. We introduce the concept of eye-naked classification, which integrates domain-specific knowledge and clinical expertise into the classification process. Our study demonstrates the effectiveness of the FBFT method, achieving impressive accuracies across multiple brain disorders using CNN-based classification. Specifically, we achieve accuracies of 99.82% for epilepsy, 95.91% for Alzheimer's disease (AD), 85.1% for murmur, and 100% for mental stress using CNN-based classification. Furthermore, in the context of naked-eye classification, we achieve accuracies of 78.6%, 71.9%, 82.7%, and 91.0% for epilepsy, AD, murmur, and mental stress, respectively. Additionally, we incorporate a mean correlation coefficient (mCC) based channel selection method to enhance the accuracy of our classification further. By combining these innovative approaches, our study enhances the visualization of EEG signals, providing medical professionals with a deeper understanding of TF medical images. This research has the potential to bridge the gap between image classification and visual medical interpretation, leading to better disease detection and improved patient care in the field of neuroscience.

A Noise-Robust Heart Sound Segmentation Algorithm Based on Shannon Energy

Heart sound segmentation has been shown to improve the performance of artificial intelligence (AI)-based auscultation decision support systems increasingly viewed as a solution to compensate for eroding auscultatory skills and the associated subjectivity. Various segmentation approaches with demonstrated performance can be utilized for this task, but their robustness can suffer in the presence of noise. A noise-robust heart sound segmentation algorithm was developed and its accuracy was tested using two datasets: the CirCor DigiScope Phonocardiogram dataset and an in-house dataset - a heart murmur library collected at the Children's National Hospital (CNH). On the CirCor dataset, our segmentation algorithm marked the boundaries of the primary heart sounds S1 and S2 with an accuracy of 0.28 ms and 0.29 ms, respectively, and correctly identified the actual positive segments with a sensitivity of 97.44%. The algorithm also executed four times faster than a logistic regression hidden semi-Markov model. On the CNH dataset, the algorithm succeeded in 87.4% cases, achieving a 6% increase in segmentation success rate demonstrated by our original Shannon energy-based algorithm. Accurate heart sound segmentation is critical to supporting and accelerating AI research in cardiovascular diseases. The proposed algorithm increases the robustness of heart sound segmentation to noise and viability for clinical use.

Markov-Based Neural Networks for Heart Sound Segmentation: Using Domain Knowledge in a Principled Way

This work considers the problem of segmenting heart sounds into their fundamental components. We unify statistical and data-driven solutions by introducing Markov-based Neural Networks (MNNs), a hybrid end-to-end framework that exploits Markov models as statistical inductive biases for an Artificial Neural Network (ANN) discriminator. We show that an MNN leveraging a simple one-dimensional Convolutional ANN significantly outperforms two recent purely data-driven solutions for this task in two publicly available datasets: PhysioNet 2016 (Sensitivity: 0.947 ±0.02; Positive Predictive Value : 0.937 ±0.025) and the CirCor DigiScope 2022 (Sensitivity: 0.950 ±0.008; Positive Predictive Value: 0.943 ±0.012). We also propose a novel gradient-based unsupervised learning algorithm that effectively makes the MNN adaptive to unseen datum sampled from unknown distributions. We perform a cross dataset analysis and show that an MNN pre-trained in the CirCor DigiScope 2022 can benefit from an average improvement of 3.90% Positive Predictive Value on unseen observations from the PhysioNet 2016 dataset using this method.

HEAR4Health: a blueprint for making computer audition a staple of modern healthcare

Recent years have seen a rapid increase in digital medicine research in an attempt to transform traditional healthcare systems to their modern, intelligent, and versatile equivalents that are adequately equipped to tackle contemporary challenges. This has led to a wave of applications that utilise AI technologies; first and foremost in the fields of medical imaging, but also in the use of wearables and other intelligent sensors. In comparison, computer audition can be seen to be lagging behind, at least in terms of commercial interest. Yet, audition has long been a staple assistant for medical practitioners, with the stethoscope being the quintessential sign of doctors around the world. Transforming this traditional technology with the use of AI entails a set of unique challenges. We categorise the advances needed in four key pillars: Hear, corresponding to the cornerstone technologies needed to analyse auditory signals in real-life conditions; Earlier, for the advances needed in computational and data efficiency; Attentively, for accounting to individual differences and handling the longitudinal nature of medical data; and, finally, Responsibly, for ensuring compliance to the ethical standards accorded to the field of medicine. Thus, we provide an overview and perspective of HEAR4Health: the sketch of a modern, ubiquitous sensing system that can bring computer audition on par with other AI technologies in the strive for improved healthcare systems.

Heart murmur detection from phonocardiogram recordings: The George B. Moody PhysioNet Challenge 2022

Cardiac auscultation is an accessible diagnostic screening tool that can help to identify patients with heart murmurs, who may need follow-up diagnostic screening and treatment for abnormal cardiac function. However, experts are needed to interpret the heart sounds, limiting the accessibility of cardiac auscultation in resource-constrained environments. Therefore, the George B. Moody PhysioNet Challenge 2022 invited teams to develop algorithmic approaches for detecting heart murmurs and abnormal cardiac function from phonocardiogram (PCG) recordings of heart sounds. For the Challenge, we sourced 5272 PCG recordings from 1452 primarily pediatric patients in rural Brazil, and we invited teams to implement diagnostic screening algorithms for detecting heart murmurs and abnormal cardiac function from the recordings. We required the participants to submit the complete training and inference code for their algorithms, improving the transparency, reproducibility, and utility of their work. We also devised an evaluation metric that considered the costs of screening, diagnosis, misdiagnosis, and treatment, allowing us to investigate the benefits of algorithmic diagnostic screening and facilitate the development of more clinically relevant algorithms. We received 779 algorithms from 87 teams during the Challenge, resulting in 53 working codebases for detecting heart murmurs and abnormal cardiac function from PCG recordings. These algorithms represent a diversity of approaches from both academia and industry, including methods that use more traditional machine learning techniques with engineered clinical and statistical features as well as methods that rely primarily on deep learning models to discover informative features. The use of heart sound recordings for identifying heart murmurs and abnormal cardiac function allowed us to explore the potential of algorithmic approaches for providing more accessible diagnostic screening in resource-constrained environments. The submission of working, open-source algorithms and the use of novel evaluation metrics supported the reproducibility, generalizability, and clinical relevance of the research from the Challenge.

Beyond Heart Murmur Detection: Automatic Murmur Grading From Phonocardiogram

OBJECTIVE

Murmurs are abnormal heart sounds, identified by experts through cardiac auscultation. The murmur grade, a quantitative measure of the murmur intensity, is strongly correlated with the patient's clinical condition. This work aims to estimate each patient's murmur grade (i.e., absent, soft, loud) from multiple auscultation location phonocardiograms (PCGs) of a large population of pediatric patients from a low-resource rural area.

METHODS

The Mel spectrogram representation of each PCG recording is given to an ensemble of 15 convolutional residual neural networks with channel-wise attention mechanisms to classify each PCG recording. The final murmur grade for each patient is derived based on the proposed decision rule and considering all estimated labels for available recordings. The proposed method is cross-validated on a dataset consisting of 3456 PCG recordings from 1007 patients using a stratified ten-fold cross-validation. Additionally, the method was tested on a hidden test set comprised of 1538 PCG recordings from 442 patients.

RESULTS

The overall cross-validation performances for patient-level murmur gradings are 86.3% and 81.6% in terms of the unweighted average of sensitivities and F1-scores, respectively. The sensitivities (and F1-scores) for absent, soft, and loud murmurs are 90.7% (93.6%), 75.8% (66.8%), and 92.3% (84.2%), respectively. On the test set, the algorithm achieves an unweighted average of sensitivities of 80.4% and an F1-score of 75.8%.

CONCLUSIONS

This study provides a potential approach for algorithmic pre-screening in low-resource settings with relatively high expert screening costs.

SIGNIFICANCE

The proposed method represents a significant step beyond detection of murmurs, providing characterization of intensity, which may provide an enhanced classification of clinical outcomes.

Early-warning of Cardiac Condition through Detection of Murmur in Heart Sound - A Case Study

Cardiovascular disease, particularly Rheumatic Heart Disease (RHD), is one of the leading causes of death in many developing countries. RHD is manageable and treatable with early detection. However, multiple countries across the globe suffer from a scarcity of experienced physicians who can perform screening at large scales. Advancements in machine learning and signal processing have paved way for Phonocardiogram (PCG)-based automatic heart sound classification. The direct implication of such methods is that it is possible to enable a person without specialized training to detect potential cardiac conditions with just a digital stethoscope. Hospitalization or life-threatening situations can be dramatically reduced via such early screenings. Towards this, we conducted a case study amongst a population from a particular geography using machine learning and deep learning methods for the detection of murmur in heart sounds. The methodology consists of first pre-processing and identifying normal vs. abnormal heart sound signals using 3 state-of-the-art methods. The second step further identifies the murmur to be systolic or diastolic by capturing the auscultation location. Abnormal findings are then sent for early attention of clinicians for proper diagnosis. The case study investigates the efficacy of the automated method employed for early screening of potential RHD and initial encouraging results of the study are presented.

An Optimal Approach for Heart Sound Classification Using Grid Search in Hyperparameter Optimization of Machine Learning

Heart-sound auscultation is one of the most widely used approaches for detecting cardiovascular disorders. Diagnosing abnormalities of heart sound using a stethoscope depends on the physician's skill and judgment. Several studies have shown promising results in automatically detecting cardiovascular disorders based on heart-sound signals. However, the accuracy performance needs to be enhanced as automated heart-sound classification aids in the early detection and prevention of the dangerous effects of cardiovascular problems. In this study, an optimal heart-sound classification method based on machine learning technologies for cardiovascular disease prediction is performed. It consists of three steps: pre-processing that sets the 5 s duration of the PhysioNet Challenge 2016 and 2022 datasets, feature extraction using Mel frequency cepstrum coefficients (MFCC), and classification using grid search for hyperparameter tuning of several classifier algorithms including k-nearest neighbor (K-NN), random forest (RF), artificial neural network (ANN), and support vector machine (SVM). The five-fold cross-validation was used to evaluate the performance of the proposed method. The best model obtained classification accuracy of 95.78% and 76.31%, which was assessed using PhysioNet Challenge 2016 and 2022, respectively. The findings demonstrate that the suggested approach obtained excellent classification results using PhysioNet Challenge 2016 and showed promising results using PhysioNet Challenge 2022. Therefore, the proposed method has been potentially developed as an additional tool to facilitate the medical practitioner in diagnosing the abnormality of the heart sound.