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.

Evaluation of Cardiac Function Characteristics after Patent Ductus Arteriosus Closure in Children and Adults by Echocardiographic Data.

This study was to investigate the value of echocardiographic data in assessing changes in cardiac function before and after transcatheter closure in children and adult patients with patent ductus arteriosus (PDA). In this study, 150 patients with isolated PDA treated by cardiac catheterization and transcatheter closure were selected as the study sample. Real-time color Doppler echocardiography was used both after and after operation. The results showed that the left ventricle returned to normal in 75 patients one day after operation, with an average age of 10.95 ± 3.27 years; the left ventricle did not return to normal in 10 patients 360 days after operation, with an average age of 64.31 ± 7.05 years. Left ventricular end diastolic volume index (LVEDVI) and left ventricular end systolic volume index (LVESVI) of patients decreased significantly one day after operation and remained at 51.95 ± 9.55 mL/m2 and 20.36 ± 8.11 mL/m-2, respectively. In summary, echocardiographic data have a high reference value in assessing cardiac function characteristics in children and adult patients with PDA and are worthy of further promotion.

Inactive Lifestyles Among Young Children With Innocent Murmurs or Congenital Heart Disease, Regardless of Disease Severity or Treatment.

BACKGROUND: Sedentary lifestyle morbidities are common among children with congenital heart disease (CHD). Understanding the physical activity trajectory from early childhood could enhance timing and effectiveness of interventions. METHODS: We recruited 154 children (56% male) at 12 to 47 months of age for this prospective, longitudinal, observational study. Physical activity and sedentary behaviour (7-day accelerometry) and motor skill (Peabody Developmental Motor Scales-2) were assessed every 8 months until 5 years of age and then annually. Mixed-effect repeated measures regression models described outcome trajectories across study assessments. RESULTS: Children had innocent heart murmurs (n = 28), CHD with insignificant hemodynamics not requiring treatment (n = 47), CHD treated by catheterization or surgery without cardiopulmonary bypass (n = 31), or CHD treated surgically with bypass (n = 48). Motor skill was age appropriate (Peabody 49.0 ± 8.4), but participants had lower physical activity (143 ± 41 minutes per day) and higher sedentary time (598 ± 89 minutes per day) than healthy peers, starting at 18 months of age. Movement behaviours were not related to treatment group (P > 0.10), and physical activity was below the recommended 180 minutes per day. Over time, physical activity, sedentary time, and motor skills were primarily related to the baseline measure of each outcome (P < 0.001). CONCLUSIONS: Children with simple or complex CHD or innocent heart murmurs have increased risk for sedentary lifestyles. Their physical activity and sedentary behaviours are established before 2 years of age, persist until school age, and are unrelated to motor skills. These results emphasize the need for interventions targeting the youngest children seen in cardiac clinics, regardless of diagnoses of CHD or innocent murmur.

Application of multi-layer denoising based on ensemble empirical mode decomposition in extraction of fault feature of rotating machinery.

Aiming at the problem that the weak features of non-stationary vibration signals are difficult to extract under strong background noise, a multi-layer noise reduction method based on ensemble empirical mode decomposition (EEMD) is proposed. First, the original vibration signal is decomposed by EEMD, and the main intrinsic modal components (IMF) are selected using comprehensive evaluation indicators; the second layer of filtering uses wavelet threshold denoising (WTD) to process the main IMF components. Finally, the virtual noise channel is introduced, and FastICA is used to de-noise and unmix the IMF components processed by the WTD. Next, perform spectral analysis on the separated useful signals to highlight the fault frequency. The feasibility of the proposed method is verified by simulation, and it is applied to the extraction of weak signals of faulty bearings and worn polycrystalline diamond compact bits. The analysis of vibration signals shows that this method can efficiently extract weak fault characteristic information of rotating machinery.

Murmurs other than the early diastolic murmur in aortic dissection.

The purpose of this review is to draw attention to the presence and significance of murmurs other than the murmur of aortic regurgitation, in patients with aortic dissection. For that purpose, a literature search was conducted using Pubmed and Googlescholar. The search terms were "dissecting aneurysm of the aorta", "systolic murmurs", "ejection systolic murmurs", "holosystolic" murmurs, "continuous murmurs", and "Austin-Flint" murmur. Murmurs other than the murmur of aortic regurgitation, which were associated with aortic dissection, fell into the categories of systolic murmurs, some of which were holosystolic, and continuous murmurs, the latter attributable to fistulae between the dissecting aneurysm and the left atrium, right atrium, and the pulmonary artery, respectively. Mid-diastolic murmurs were also identified, and these typically occurred in association with both the systolic and the early diastolic murmurs. Among patients with systolic murmurs clinical features which enhanced the pre-test probability of aortic dissection included back pain, stroke, paraplegia, unilateral absence of pulses, interarm differences in blood pressure, hypertension, shock, bicuspid aortic valve, aortic coarctation, Turner's syndrome, and high D-dimer levels, respectively. In the absence of the murmur of aortic regurgitation timely diagnosis of aortic dissection could be expedited by increased attention to parameters which enhance pretest probability of aortic dissection. That logic would apply even if the only murmurs which were elicited were systolic murmurs.

S-Nitrosylation of Histone Deacetylase 2 by Neuronal Nitric Oxide Synthase as a Mechanism of Diastolic Dysfunction.

BACKGROUND: Although the clinical importance of heart failure with preserved ejection fraction has been extensively explored, most therapeutic regimens, including nitric oxide (NO) donors, lack therapeutic benefit. Although the clinical characteristics of heart failure with preserved ejection fraction are somewhat heterogeneous, diastolic dysfunction (DD) is one of the most important features. Here we report that neuronal NO synthase (nNOS) induces DD by S-nitrosylation of HDAC2 (histone deacetylase 2). METHODS: Two animal models of DD-SAUNA (SAlty drinking water/Unilateral Nephrectomy/Aldosterone) and mild transverse aortic constriction mice-as well as human heart samples from patients with left ventricular hypertrophy were used. Genetically modified mice that were either nNOS-ablated or HDAC2 S-nitrosylation-resistant were also challenged. N(ω)-propyl-L-arginine, an nNOS selective inhibitor, and dimethyl fumarate, an NRF2 (nuclear factor erythroid 2-related factor 2) inducer, were used. Molecular events were further checked in human left ventricle specimens. RESULTS: SAUNA or mild transverse aortic constriction stress impaired diastolic function and exercise tolerance without overt systolic failure. Among the posttranslational modifications tested, S-nitrosylation was most dramatically increased in both models. Utilizing heart samples from both mice and humans, we observed increases in nNOS expression and NO production. N(ω)-propyl-L-arginine alleviated the development of DD in vivo. Similarly, nNOS knockout mice were resistant to SAUNA stress. nNOS-induced S-nitrosylation of HDAC2 was relayed by transnitrosylation of GAPDH. HDAC2 S-nitrosylation was confirmed in both DD mouse and human left ventricular hypertrophy. S-nitrosylation of HDAC2 took place at C262 and C274. When DD was induced, HDAC2 S-nitrosylation was detected in wild-type mouse, but not in HDAC2 knock-in mouse heart that expressed HDAC2 C262A/C274A. In addition, HDAC2 C262A/C274A mice maintained normal diastolic function under DD stimuli. Gene delivery with adenovirus-associated virus 9 (AAV9)-NRF2, a putative denitrosylase of HDAC2, or pharmacological intervention by dimethyl fumarate successfully induced HDAC2 denitrosylation and mitigated DD in vivo. CONCLUSIONS: Our observations are the first to demonstrate a new mechanism underlying DD pathophysiology. Our results provide theoretical and experimental evidence to explain the ineffectiveness of conventional NO enhancement trials for improving DD with heart failure symptoms. More important, our results suggest that reduction of NO or denitrosylation of HDAC2 may provide a new therapeutic platform for the treatment of refractory heart failure with preserved ejection fraction.

Identifying cardiogenic shock in the emergency department.

INTRODUCTION: Cardiogenic shock is difficult to diagnose due to diverse presentations, overlap with other shock states (i.e. sepsis), poorly understood pathophysiology, complex and multifactorial causes, and varied hemodynamic parameters. Despite advances in interventions, mortality in patients with cardiogenic shock remains high. Emergency clinicians must be ready to recognize and start appropriate therapy for cardiogenic shock early. OBJECTIVE: This review will discuss the clinical evaluation and diagnosis of cardiogenic shock in the emergency department with a focus on the emergency clinician. DISCUSSION: The most common cause of cardiogenic shock is a myocardial infarction, though many causes exist. It is classically diagnosed by invasive hemodynamic measures, but the diagnosis can be made in the emergency department by clinical evaluation, diagnostic studies, and ultrasound. Early recognition and stabilization improve morbidity and mortality. This review will focus on identification of cardiogenic shock through clinical examination, laboratory studies, and point-of-care ultrasound. CONCLUSIONS: The emergency clinician should use the clinical examination, laboratory studies, electrocardiogram, and point-of-care ultrasound to aid in the identification of cardiogenic shock. Cardiogenic shock has the potential for significant morbidity and mortality if not recognized early.

Coronary artery fistula in adults: Incidence and appearance on cardiac computed tomography and comparison of detectability and hemodynamic effects with those on transthoracic echocardiography.

BACKGROUND: The widespread use of cardiac computed tomography (CT) has increased the incidental discovery of fistulas of the coronary artery (CAF). This condition is rare and can affect hemodynamic parameters, but few reports focus on its hemodynamic effects. We investigated the frequency and types of CAF on CT and compared them with those of transthoracic echocardiography (TTE) to evaluate the detectability of CAF and its hemodynamic effects. MATERIALS AND METHODS: We retrospectively evaluated cardiac CT images of 6789 adult patients who underwent imaging from January 1, 2013 through September 30, 2019 at our institution. We assessed the CT images for the presence of CAF and compared our findings with those obtained by TTE in control cases without CAF. RESULTS: The prevalence of CAF determined with cardiac CT was 0.91%, with the left anterior descending artery (67.7%) as the most common site of origin and the main pulmonary artery (82.3%) as the most common origin of drainage. The incidence of aneurysm accompanying CAF was 48.4%. Color Doppler in TTE demonstrated abnormal flow that would suggest the presence of CAF of only 23.1%. Echocardiographic findings of hemodynamics did not differ significantly between patients with and without CAF nor between elderly and non-elderly patients with CAF. CONCLUSIONS: Our study revealed differing prevalence and types of CAF from those reported using coronary angiography and little impact of CAF on hemodynamics. Color Doppler in TTE did not readily depict abnormal flow that might suggest the presence of a fistula, but cardiac CT allowed noninvasive and comprehensive assessment of CAF. Thus, we believe the acquisition of cardiac CT is necessary to establish the cause of continuous murmur that is not identified with TTE.

Emergency caesarean section of a patient with eisenmenger's syndrome: A tight-rope walk.

Successful management of a pregnant patient with complex congenital heart disease is a challenge for anesthesiologists, requiring thorough knowledge of the impact of pregnancy on the cardiac lesion. Hearing and speech impaired patients pose a barrier to effective communication between the patient and the doctors, thus increasing the anxiety and risk of complications. Here, we present a case of a hearing and speech impaired woman with the rare and dangerous Eisenmenger's syndrome, presenting for an emergency cesarean section (CS).

Does Echocardiography Have a Role in the Cardiologist's Diagnosis of Innocent Murmurs in Childhood?

BACKGROUND: Innocent murmurs are common in childhood. Echocardiography may diagnose almost all congenital cardiac abnormalities. Earlier studies have suggested that it may be unnecessary when a clinical diagnosis of an innocent murmur is made by an experienced cardiologist. Yet, echocardiography is commonly performed despite such a diagnosis. Is that justifiable? METHODS: Patients referred to a paediatric cardiologist for evaluation of a murmur were assessed. If an innocent murmur was clinically diagnosed, an echocardiogram and an electrocardiogram were ordered when requested. The cardiologist completed a questionnaire documenting the reasons for the investigations and his degree of confidence in the clinical diagnosis. The parents and patients were informed of the diagnosis and the results of the investigations. RESULTS: In almost all 62 cases whose murmurs were clinically considered to be innocent, the referring doctor expected an echocardiogram to be performed. Following echocardiography, one patient was found to have a small muscular ventricular septal defect. On review, the murmur was consistent with that diagnosis. Three (3) others had abnormal echocardiograms namely congenitally corrected transposition of great arteries, partial anomalous right upper lobe pulmonary vein, and left ventricular non-compaction. All on review were still considered clinically to have innocent murmurs. CONCLUSIONS: Despite the competency of the cardiologist, echocardiography diagnosed important but relatively uncommon cardiac abnormalities with prognostic implications that would have been missed if only the clinical examination was performed. Over and above the referring doctor's expectations for an echocardiogram to be performed is the need for the cardiologist to provide a complete and definitive diagnosis.

Towards classifying non-segmented heart sound records using instantaneous frequency based features.

Heart sound and its recorded signal which is known as phonocardiograph (PCG) are one of the most important biosignals that can be used to diagnose cardiac diseases alongside electrocardiogram (ECG). Over the past few years, the use of PCG signals has become more widespread and researchers pay their attention to it and aim to provide an automated heart sound analysis and classification system that supports medical professionals in their decision. In this paper, a new method for heart sound features extraction for the classification of non-segmented signals using instantaneous frequency was proposed. The method has two major phases: the first phase is to estimate the instantaneous frequency of the recorded signal; the second phase is to extract a set of eleven features from the estimated instantaneous frequency. The method was tested into two different datasets, one for binary classification (Normal and Abnormal) and the other for multi-classification (Five Classes) to ensure the robustness of the extracted features. The overall accuracy, sensitivity, specificity, and precision for binary classification and multi-classification were all above 95% using both random forest and KNN classifiers.

Effect of pimobendan on physical fitness, lactate and echocardiographic parameters in dogs with preclinical mitral valve disease without cardiomegaly.

Pimobendan has gained enormous importance in the treatment of mitral valve disease in dogs. The current consensus statement of the American College of Veterinary Internal Medicine (ACVIM) recommends a treatment for dogs with symptomatic disease and dogs with asymptomatic disease with radiographic and echocardiographic signs of cardiomegaly. To investigate whether these dogs also benefit from a therapy with pimobendan, 21 dogs with mitral valve disease ACVIM B1 underwent a standardized submaximal exercise test on a treadmill. In this double-blinded and randomized study, the animals were divided into two groups, one receiving pimobendan and the other a placebo. At the first visit and at every follow-up appointment (at days 90 and 180), heart rate during the complete exercise test and lactate before and after running were measured. In addition to this, a questionnaire was completed by the dogs' owners and all dogs were given an echocardiographic examination to detect any changes and to observe if the disease had progressed. Due to the diagnosis of leishmaniosis, one dog in the pimobendan group was retrospectively removed from the study so that 20 dogs were included for statistical analysis. No differences were observed at any time between the pimobendan-group and the placebo-group regarding heart rate. At day 180, the increase in lactate after exercise was significantly lower than in the placebo-group. The increase in the pimobendan-group at day 180 was lower than at day 90. Most of the dog owners from the pimobendan-group declared that their dogs were more active at day 90 (6/10) and at day 180 (8/10), while most dog owners from the placebo-group observed no changes regarding activity at day 90 (8/10) and day 180 (6/10). It can be concluded that the results of this study indicate that some dogs with mitral valve disease ACVIM B1 might benefit from a therapy with pimobendan.

An efficient heart murmur recognition and cardiovascular disorders classification system.

The problem addressed in this work is the detection of a heart murmur and the classification of the associated cardiovascular disorder based on the heart sound signal. For this purpose, a dataset of Phonocardiogram (PCG) signals is acquired using baseline conditions. The dataset is acquired from 283 volunteers using Littman 3200 electronic stethoscope for a normal and four different types of heart murmurs. The samples are labelled and validated through echocardiography test of each participating volunteer. For feature extraction, normalized average Shannon energy with time-domain characteristics of heart sound signal is exploited to segment the PCG signal into its components. To improve the quality of the features, in contrast to the previous methods, all systole and diastole intervals are utilized to extract 50 Mel-Frequency Cepstrum Coefficients (MFCC) based features. Then, the iterative backward elimination method is used to identify and remove the redundant features to reduce the complexity in order to conceive a computationally tractable system. An MFCC feature vector of dimension 26 is selected for training seven different types of Support Vector Machine (SVM) and K-Nearest Neighbors (KNN) based classifiers for detection and classification of cardiovascular disorders. Fivefold cross-validation and 20% data holdout validation schemes are used for testing the classifiers. Classification accuracy of 92.6% is achieved using selected features and medium Gaussian SVM classifier. The learning curves show a good bias-variance trade-off indicating a well-fitted and generalized model for making future predictions.

Age when presumptive innocent cardiac murmurs spontaneously disappear in clinically healthy Cairn terrier puppies.

Innocent cardiac murmurs are common findings in Cairn terrier puppies during their first veterinary health check. The age when these murmurs spontaneously disappear is unreported. The purpose of this study was to establish the age when presumably innocent cardiac murmurs disappear in a population of clinically healthy Cairn terrier puppies. Over a period of 9 months, 227 clinically healthy Cairn terrier puppies (median age, 53 days; range, 45-76 days) underwent auscultation by a veterinary cardiology specialist, who identified 82 puppies with a presumably innocent cardiac murmur. Owners of 20 puppies volunteered to return to the clinic for serial rechecks. Owners of three puppies returned only once, therefore these puppies were censored. Hence this longitudinal observational study followed 17 puppies with monthly cardiac auscultation until the disappearance of their murmurs. The median age of the 20 puppies when the murmur was last audible was 65 days (range 52-285 days). The median age of the 17 puppies when the murmur was no longer audible was 87 days (range 71-347 days; 95% confidence interval 63-111 days). Four of the 17 puppies had a murmur after 3 months of age and two of them had a murmur beyond 6 months of age. The oldest puppy to have an audible murmur was 9.5 months old; this murmur was not audible at 11.5 months of age. In most Cairn terrier puppies in this population, the presumably innocent murmur resolved spontaneously by 3 months of age.

Assessment of the Cardiovascular System in Horses During Prepurchase and Insurance Examinations.

Arrhythmias detected on prepurchase examination should be confirmed with an ECG. Exercising ECG determines if the arrhythmia is overdriven during exercise or is a safety concern. An echocardiogram is needed in all horses with a grade 3/6 or louder mid to late systolic, holosystolic, or pansystolic murmur or any holodiastolic decrescendo murmur to identify the cardiac abnormality and its hemodynamic impact. Most horses with arrhythmias and murmurs have a normal performance career and life expectancy and are insurable. Risks for sudden death and congestive heart failure associated with the common murmurs and arrhythmias are identified, because these horses cannot be insured.