-
IEEE Journal of Biomedical and Health... Apr 2024One in every four newborns suffers from congenital heart disease (CHD) that causes defects in the heart structure. The current gold-standard assessment technique,...
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.
Topics: Adolescent; Child; Humans; Infant, Newborn; Heart Auscultation; Deep Learning; Heart Murmurs; Phonocardiography; Auscultation; Heart Defects, Congenital
PubMed: 38261492
DOI: 10.1109/JBHI.2024.3357506 -
Acta Paediatrica (Oslo, Norway : 1992) Jan 2024Our aim was to assess undiagnosed congenital heart defects (CHD) after newborns' hospital discharge in patients with a murmur or CHD suspicion, to find out the signs... (Review)
Review
AIM
Our aim was to assess undiagnosed congenital heart defects (CHD) after newborns' hospital discharge in patients with a murmur or CHD suspicion, to find out the signs that predict CHDs and to estimate the costs of the examinations.
METHODS
We reviewed retrospective medical records of patients (n = 490) referred for the evaluation of CHD suspicion during 2017-2018.
RESULTS
The median age of the patients was 2.5 (IQR 0.5-7.4) years. Sixty-three (13%) patients had an abnormal echocardiography. Neither ductal-dependent nor cyanotic CHDs were found. Cardiac interventions were performed for 14 out of 63 (22%) patients. Clinical signs indicating CHDs were murmur grade ≥3 (10/11 [91%] vs. 53/479 [11%], p < 0.001) and harsh murmur (15/44 [34%] vs. 48/446 [11%], p < 0.001). Abnormal electrocardiography did not indicate CHD (8/40 [20%] vs. 55/447 [12%], p = 0.165). The total cost of the examinations was 259 700€. The share of the cost of studies assessed as benign was 59%.
CONCLUSION
Only a few CHDs were found after newborn hospital discharge among patients who received foetal and newborn screening and were examined due to CHD suspicion. The high number of benign murmurs in children leads to many referrals, resulting in unnecessary healthcare costs.
Topics: Child; Humans; Infant, Newborn; Infant; Child, Preschool; Patient Discharge; Retrospective Studies; Heart Defects, Congenital; Heart Murmurs; Hospitals
PubMed: 37522553
DOI: 10.1111/apa.16928 -
Animals : An Open Access Journal From... Jan 2024Mitral and aortic valve insufficiencies have been commonly reported in horses. The objective of this study was to establish the use of acoustic cardiography (Audicor) in...
Mitral and aortic valve insufficiencies have been commonly reported in horses. The objective of this study was to establish the use of acoustic cardiography (Audicor) in horses with aortic (AI) or mitral valve insufficiency (MI). A total of 17 healthy horses, 18 horses with AI, and 28 horses with MI were prospectively included. None of the horses was in heart failure. Echocardiography and Audicor analyses were conducted. Electromechanical activating time (EMAT), rate-corrected EMATc, left ventricular systolic time (LVST), rate-corrected LVSTc, and intensity and persistence of the third and fourth heart sound (S3, S4) were reported by Audicor. Graphical analysis of the three-dimensional (3D) phonocardiogram served to visually detect murmurs. Audicor snapshot variables were compared between groups using one-way ANOVA followed by Tukey's multiple-comparisons test. The association between Audicor snapshot variables and the corresponding echocardiographic variables was investigated by linear regression and Bland-Altman analyses. Heart murmurs were not displayed on Audicor phonocardiograms. No significant differences were found between Audicor variables obtained in clinically healthy horses and horses with valvular insufficiency. The Audicor device is unable to detect heart murmurs in horses. Audicor variables representing cardiac function are not markedly altered, and their association with corresponding echocardiographic variables is poor in horses with valvular insufficiency that are not in heart failure.
PubMed: 38275790
DOI: 10.3390/ani14020331 -
Sensors (Basel, Switzerland) Jun 2024The phonocardiogram (PCG) can be used as an affordable way to monitor heart conditions. This study proposes the training and testing of several classifiers based on SVMs...
The phonocardiogram (PCG) can be used as an affordable way to monitor heart conditions. This study proposes the training and testing of several classifiers based on SVMs (support vector machines), k-NN (k-Nearest Neighbor), and NNs (neural networks) to perform binary ("Normal"/"Pathologic") and multiclass ("Normal", "CAD" (coronary artery disease), "MVP" (mitral valve prolapse), and "Benign" (benign murmurs)) classification of PCG signals, without heart sound segmentation algorithms. Two datasets of 482 and 826 PCG signals from the Physionet/CinC 2016 dataset are used to train the binary and multiclass classifiers, respectively. Each PCG signal is pre-processed, with spike removal, denoising, filtering, and normalization; afterward, it is divided into 5 s frames with a 1 s shift. Subsequently, a feature set is extracted from each frame to train and test the binary and multiclass classifiers. Concerning the binary classification, the trained classifiers yielded accuracies ranging from 92.4 to 98.7% on the test set, with memory occupations from 92.7 kB to 11.1 MB. Regarding the multiclass classification, the trained classifiers achieved accuracies spanning from 95.3 to 98.6% on the test set, occupying a memory portion from 233 kB to 14.1 MB. The NNs trained and tested in this work offer the best trade-off between performance and memory occupation, whereas the trained k-NN models obtained the best performance at the cost of large memory occupation (up to 14.1 MB). The classifiers' performance slightly depends on the signal quality, since a denoising step is performed during pre-processing. To this end, the signal-to-noise ratio (SNR) was acquired before and after the denoising, indicating an improvement between 15 and 30 dB. The trained and tested models occupy relatively little memory, enabling their implementation in resource-limited systems.
Topics: Humans; Phonocardiography; Machine Learning; Signal Processing, Computer-Assisted; Algorithms; Neural Networks, Computer; Wearable Electronic Devices; Support Vector Machine
PubMed: 38931636
DOI: 10.3390/s24123853 -
International Journal of Cardiology.... Apr 2024Insufficient clinicians' auscultation ability delays the diagnosis and treatment of valvular heart disease (VHD); artificial intelligence provides a solution to...
BACKGROUND
Insufficient clinicians' auscultation ability delays the diagnosis and treatment of valvular heart disease (VHD); artificial intelligence provides a solution to compensate for the insufficiency in auscultation ability by distinguishing between heart murmurs and normal heart sounds. However, whether artificial intelligence can automatically diagnose VHD remains unknown. Our objective was to use deep learning to process and compare raw heart sound data to identify patients with VHD requiring intervention.
METHODS
Heart sounds from patients with VHD and healthy controls were collected using an electronic stethoscope. Echocardiographic findings were used as the gold standard for this study. According to the chronological order of enrollment, the early-enrolled samples were used to train the deep learning model, and the late-enrollment samples were used to validate the results.
RESULTS
The final study population comprised 499 patients (354 in the algorithm training group and 145 in the result validation group). The sensitivity, specificity, and accuracy of the deep-learning model for identifying various VHDs ranged from 71.4 to 100.0%, 83.5-100.0%, and 84.1-100.0%, respectively; the best diagnostic performance was observed for mitral stenosis, with a sensitivity of 100.0% (31.0-100.0%), a specificity of 100% (96.7-100.0%), and an accuracy of 100% (97.5-100.0%).
CONCLUSIONS
Based on raw heart sound data, the deep learning model effectively identifies patients with various types of VHD who require intervention and assists in the screening, diagnosis, and follow-up of VHD.
PubMed: 38482387
DOI: 10.1016/j.ijcha.2024.101368 -
Pediatric Emergency Care Oct 2023Diagnosis of acute myocarditis or dilated cardiomyopathy (DCM) on initial presentation is difficult in children younger than 2 years because most present with complaints...
OBJECTIVE
Diagnosis of acute myocarditis or dilated cardiomyopathy (DCM) on initial presentation is difficult in children younger than 2 years because most present with complaints suggestive of a respiratory infection. The objective of this study is to determine whether signs, symptoms, and diagnostic studies excluding those of heart failure, done routinely in the emergency department could distinguish children younger than 2 years with acute myocarditis or DCM from those with respiratory illnesses.
METHODS
Sixty-four infants' charts, 32 cases and 32 controls, were reviewed from January 1, 2009, through December 31, 2020. Controls were matched to cases with respect to age, reason, and time of admission. Signs, symptoms, and blood gases were reviewed.
RESULTS
The median age is 6.5 (0.5-22) months in both groups. Infants presenting with signs of heart failure including murmurs ( P = 0.002), prolonged capillary refill ( P = 0.024), cool, mottled extremities ( P = 0.002), poor perfusion ( P = 0.001), or hepatomegaly ( P < 0.001) were more likely to be diagnosed with acute myocarditis or DCM when compared with the control group with respiratory disease. Infants with fever ( P = 0.017), nasal congestion ( P < 0.001), rhinorrhea ( P < 0.001), cough ( P < 0.001), and wheezing ( P < 0.001) were more likely to have a respiratory illness than acute myocarditis or DCM. The presence of a lower p co2 (30 [14-116] vs 40 [31-59] mm Hg, P < 0.001), lower bicarbonate (16.7 [6.3-23.4] vs 21.7 [16-28.4], P < 0.001), or an oxygen saturation > 95% ( P = 0.004) was observed in infants with acute myocarditis or DCM compared with those with respiratory illness. By multivariable analysis, infants with tachycardia in the absence of fever, metabolic acidosis, and an oxygen saturation > 95% were more likely to have acute myocarditis or DCM than those without this disease.
CONCLUSIONS
Children younger than 2 years presenting to the emergency department with tachycardia and no fever, metabolic acidosis, and a high oxygen saturation should be investigated for acute myocarditis or DCM.
Topics: Infant; Child; Humans; Myocarditis; Cardiomyopathy, Dilated; Heart Failure; Early Diagnosis
PubMed: 37590924
DOI: 10.1097/PEC.0000000000003038 -
American Journal of Veterinary Research Jan 2024To develop breed-specific echocardiographic values for normal Borzoi and to report the prevalence of structural cardiac abnormalities.
OBJECTIVE
To develop breed-specific echocardiographic values for normal Borzoi and to report the prevalence of structural cardiac abnormalities.
ANIMALS
146 clinically healthy, adult Borzoi dogs.
METHODS
Cardiac auscultation and standard echocardiograms were performed. Longitudinal follow-up was described in a subset of dogs (n = 25).
RESULTS
Most Borzoi were structurally normal (119/146, 81.5%), with breed-specific echocardiographic values generated independently for each sex, as females weighed significantly less than males (30.4 ± 3.8 kg vs 38.3 ± 4.1 kg, respectively; P < .001), and a significant impact of sex was found on most measurements. Physiologic heart murmurs were identified in 64/119 (53.8%) normal dogs. Thirty-six (30.2%) structurally normal dogs had trace or mild mitral regurgitation, and 43 (36.1%) had trace or mild tricuspid regurgitation. Structural cardiac disease was identified in 21 dogs (14.4%), including 9 dogs (6.2%) with dilated cardiomyopathy (DCM), 9 dogs (6.2%) with stage B1 myxomatous mitral valve disease (MMVD), and 3 (2.1%) dogs with congenital abnormalities. Seven dogs (4.8%) had equivocal abnormalities. During follow-up, new dogs were diagnosed with occult DCM (n = 3), equivocal DCM (1), and stage B1 MMVD (2). Two dogs originally diagnosed with DCM (1 occult and 1 equivocal) normalized after diet change.
CLINICAL RELEVANCE
Borzoi dogs commonly have physiologic heart murmurs and mild atrioventricular valve regurgitation. Both DCM and MMVD were identified at similar frequencies in healthy Borzoi, although dogs with MMVD all had normal heart sizes. Echocardiographic screening for DCM in Borzoi should be considered, with breed-specific echocardiographic values now available for improved diagnostic confidence.
PubMed: 38154250
DOI: 10.2460/ajvr.23.11.0255 -
Journal of the American Veterinary... Nov 2023To retrospectively evaluate neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) as a biomarker for severity and short-term outcomes of congestive...
Neutrophil-to-lymphocyte ratio is increased in dogs with acute congestive heart failure secondary to myxomatous mitral valve disease compared to both dogs with heart murmurs and healthy controls.
OBJECTIVE
To retrospectively evaluate neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) as a biomarker for severity and short-term outcomes of congestive heart failure (CHF) secondary to myxomatous mitral valve disease (MMVD) in dogs.
ANIMALS
47 dogs with CHF secondary to MMVD, 47 dogs with presumptive preclinical MMVD, and 47 control dogs.
METHODS
Medical record data (signalment, physical examination findings, medical treatments instituted, American College of Veterinary Internal Medicine MMVD stage, length of hospitalization, outcome, and hospital re-presentation due to CHF) from March 2012 through March 2022 for each dog were collected. Statistical analyses were performed with Mann-Whitney, Spearman correlation, and Fisher exact tests.
RESULTS
NLR (but not PLR) was significantly higher in dogs with CHF secondary to MMVD (6.41) compared to presumptive preclinical MMVD dogs (4.66; P < .001) and control dogs (3.95; P < .001). Dogs with higher NLR and PLR received significantly higher cumulative dosages of loop-diuretic therapy during hospitalization (ρ = 0.3, P = .04; and ρ = 0.4, P = .02, respectively). There was a positive association between NLR and duration of oxygen supplementation within the CHF group (ρ = 0.4; P = .01).
CLINICAL RELEVANCE
The increased diuretic dose and time receiving oxygen supplementation may represent increased disease severity for which NLR (and to a lesser extent PLR) may serve as a readily available marker. The data presented provide information regarding some of the systemic inflammatory changes seen in CHF secondary to MMVD in dogs. Future research should include prospective, longitudinal studies to provide insight into the long-term prognostic value of NLR and PLR in dogs with CHF.
Topics: Humans; Dogs; Animals; Mitral Valve; Retrospective Studies; Prospective Studies; Neutrophils; Heart Valve Diseases; Heart Failure; Heart Murmurs; Dog Diseases; Diuretics
PubMed: 37406992
DOI: 10.2460/javma.23.03.0131 -
Circulation. Heart Failure Oct 2023
Topics: Humans; Mitral Valve Insufficiency; Heart Failure; Mitral Valve; Mitral Valve Stenosis; Arteriovenous Fistula; Cardiac Catheterization; Hemodynamics
PubMed: 37435745
DOI: 10.1161/CIRCHEARTFAILURE.123.010733 -
PloS One 2024Transthoracic Echocardiography is the first-line, non-invasive, and accessible imaging modality to evaluate heart disease anatomy, physiology, and hemodynamics. We aim...
Two-Dimensional and Doppler trans-thoracic echocardiographic patterns of suspected pediatric heart diseases at Tibebe--Ghion specialized Teaching Hospital and Adinas General Hospital, Bahir Dar, North-west Ethiopia:-An experience from an LMIC.
BACKGROUND
Transthoracic Echocardiography is the first-line, non-invasive, and accessible imaging modality to evaluate heart disease anatomy, physiology, and hemodynamics. We aim to describe the trans-thoracic echocardiography pattern of pediatric heart diseases and reasons for referral in children referred to Bahir Dar University Tibebe-Ghion Hospital and Adinas General Hospital.
METHOD
A descriptive cross-sectional study of the archived Transthoracic, Two Dimensional, and Doppler Echocardiography assessments of children from birth to fifteen years of age performed between June 2019 to May 2023 was done. Data were collected retrospectively from February 01, 2023 -May 31, 2023. Categorical variables like gender, referral reasons for echocardiography, and patterns of pediatric heart lesions were analyzed in the form of proportions and presented in tables and figures. Discrete variables including age were summarized as means (SD) and medians(IQR).
RESULTS
Out of 3,647 Children enrolled; 1,917 (52.6%) were males and 1,730 (47.4%) were females. The median (IQR) age of children enrolled was 24 months (5 to 96). Cardiac murmur (33%) was the most common reason for echocardiography followed by, Respiratory Distress (18%), Syndromic Child (15%), easy fatigability/ Diaphoresis (14.3%), congestive heart failure (14%), and rheumatic fever (13.2%). Congenital heart defect (CHD) accounts for 70% of all heart diseases, followed by rheumatic heart disease (21%). Isolated ventricular septal defect(VSD) was the most common CHD (21%) followed by isolated Patent ductus arteriosus (15%), isolated atrial septal defect (10%), Isolated atrioventricular septal defect (6%) and isolated pulmonary stenosis (5%). Cyanotic CHD accounts for 11.5% of all heart diseases. Tetralogy of Fallot (30%), d-TGA (20%), and double outlet right ventricle (19%) were the most common cyanotic CHDs.
CONCLUSIONS
In our study, congenital heart lesions are the most common diagnosis and cardiac murmurs are the most common presenting reasons for echocardiography evaluation.
Topics: Male; Female; Child; Humans; Child, Preschool; Retrospective Studies; Cross-Sectional Studies; Developing Countries; Ethiopia; Hospitals, General; Heart Defects, Congenital; Echocardiography, Doppler; Heart Septal Defects, Ventricular; Echocardiography; Hospitals, University; Heart Murmurs
PubMed: 38466681
DOI: 10.1371/journal.pone.0292694