Objective:To evaluate the clinical utility of combining artificial intelligence (AI) with handheld focused cardiac ultrasound (FoCUS) performed by noncardiologist physicians in clinical care settings. Patients and Methods:In this prospective, single-arm study conducted from July 1, 2022, through December 31, 2023 (ClinicalTrials.gov NCT05455541), 660 adult patients presenting to the emergency department or internal medicine wards were assessed with handheld ultrasound devices enhanced by AI algorithms. These algorithms provided automated analysis of ventricular function, valvular disease, pericardial effusion, and inferior vena cava size. Participating physicians received focused training and performed examinations either in response to clinical suspicion or as part of routine evaluation. The primary outcome was whether AI-guided FoCUS contributed to new diagnoses, treatment modifications, or additional procedures. Results:Artificial intelligence-enhanced FoCUS identified clinically relevant cardiac findings in 193 patients (29%), including newly recognized valvular abnormalities and reduced left ventricular function. In 49 patients (7%), medical therapy was adjusted based on findings, and 9 patients (1.4%) underwent interventional procedures. Diagnostic performance analyses showed high sensitivity for detecting reduced left ventricular function and valvular disease, with lower sensitivity for right-sided abnormalities. Conclusion:This study demonstrates that integrating AI-enhanced FoCUS into routine workflows can provide clinically relevant information that may influence diagnostic assessment and management by noncardiology practitioners in acute care settings.

BACKGROUND:Body mass index (BMI) is an independent, predictor of adverse outcomes among patients with cardiovascular diseases. Our aim was to evaluate how BMI modifies the association of severe tricuspid regurgitation (TR) with poor survival. METHODS:Consecutive echocardiographic reports linked to clinical data from a tertiary medical center (2007-2021) were reviewed. BMI was stratified to underweight, normal weight, overweight, and obesity with cutoffs of 18.5, 25, and 30 Kg/m2. Multivariable Cox regression models were applied to calculate adjusted hazard ratios (aHR) for all-cause mortality. RESULTS:The study population included 111,830 patients with a median age of 65 (IQR 52-76) years (58% males), 3436 (3.1%) patients had severe TR. There were 2398 (2%), 40,393 (36%), 43,790 (39%), and 25,249 (24%) individuals with underweight, normal weight, overweight, and obesity, respectively. During a median follow-up of 6 (3-10) years, 29,921 (27%) patients died. There was an interaction between BMI and TR with the outcome (p < 0.01). Severe TR was associated with mortality for patients with normal weight, overweight, and obesity, with an HR of 1.29 (95% CI 1.20-1.39), 1.26 (1.16-1.37), and 1.28 (1.15-1.43), respectively. This association was not evident among patients with underweight with an HR of 0.88 (0.64-1.22). Sub-analysis of hospitalized patients, with adjustment of the model to baseline comorbidities, including frailty, yielded similar point estimates to the main analysis. CONCLUSIONS:The association of severe TR with mortality is modified by BMI group. Severe TR is associated with mortality among patients of all BMI groups, except for individuals with underweight.

BackgroundValvular heart disease and heart failure are major global health burdens, yet access to comprehensive echocardiography is often limited, particularly in resource-constrained settings. Artificial intelligence (AI) may enable rapid, point-of-care cardiac assessment using simplified imaging protocols.ObjectivesTo evaluate whether a deep learning model can accurately detect significant valvular and ventricular dysfunction using only a single two-dimensional apical four-chamber echocardiographic view, including images acquired by non-cardiologists with handheld ultrasound devices.MethodsWe retrospectively analyzed 120,127 echocardiographic studies from a tertiary medical center to train and validate a deep learning model for identifying moderate-or-greater mitral or tricuspid regurgitation, right ventricular dysfunction, and reduced left ventricular ejection fraction (≤40%). A prospective cohort of 209 patients underwent handheld point-of-care cardiac ultrasound performed by non-cardiologist physicians, with same-hospitalization comprehensive echocardiography as the reference standard.ResultsIn retrospective testing, model areas under the curve (AUCs) were 0.883 for mitral regurgitation, 0.913 for tricuspid regurgitation, 0.940 for right ventricular dysfunction, and 0.982 for reduced ejection fraction. In the prospective cohort, AUCs were 0.72, 0.87, 0.95, and 0.97 for the same respective targets.ConclusionsA single-view deep learning model demonstrated strong diagnostic accuracy for detecting significant valvular and ventricular dysfunction across both standard and handheld ultrasound acquisitions. This approach may facilitate rapid, scalable cardiac function screening by non-cardiologists in diverse clinical environments.Clinical Trial Registrationidentifier NCT05455541.

Background and objective: Artificial intelligence (AI) based analysis of electrocardiogram (ECG) signals has become a powerful tool for predicting clinically important attributes such as sex and cardiac function. The growing use of wearable ECG technologies has increased interest in optimizing deep learning models for rapid and efficient predictions. Convolutional neural networks (CNNs) are the most established architecture in this field, while Vision Transformers (ViTs), successful in image analysis, remain less studied for ECG time-series data. We aimed to compare the performance of ViTs and CNNs in predicting sex and left ventricular dysfunction (LVD), defined as ejection fraction <= 35 %, and to evaluate reduced-input models using short ECG segments and single-lead recordings. Methods: This retrospective study analyzed 12-lead ECGs recorded between January 2005 and December 2022 at a tertiary medical center. The dataset included 150,691 patients for sex classification and 29,422 for LVD. Binary classification models were developed using full-length ECGs, truncated segments (1-10 s), and individual leads. CNNs were implemented using PyTorch and ViTs via the HuggingFace framework. Model performance was assessed using area under the receiver operating characteristic curve (AUROC), with statistical comparisons by DeLong's test. Results: CNNs outperformed ViTs in both tasks. For sex prediction, AUROC was 0.915 (95 % CI: 0.912-0.918) for CNNs and 0.898 (95 % CI: 0.894-0.901) for ViTs (P < 0.001). For LVD, CNNs achieved an AUROC of 0.895 (95 % CI: 0.884-0.906) versus 0.866 (95 % CI: 0.853-0.879) for ViTs (P = 0.023). Over 98 % of peak performance was retained using 2-s segments, similarly single-lead inputs achieved a relatively high performance. Conclusion: CNN-based models demonstrated higher accuracy than ViTs for ECG-based prediction of sex and LVD. Compact input models may enable efficient clinical use. Further studies should explore where ViTs may offer added value.

[This corrects the article DOI: 10.3389/fdgth.2025.1684933.].

Aims:Biological age is increasingly recognized as a superior predictor of morbidity, mortality, compared with chronological age. Artificial intelligence (AI)-driven ageing clocks enable rapid, non-invasive assessment. Cardiovascular (CV) ageing is of particular relevance given its central role in systemic metabolic health. This study evaluated the clinical utility of an ultrasound (US)-based CV biological age clock derived from handheld point-of-care ultrasound (POCUS), in comparison with haematological and electrocardiographic (ECG)-based clocks. Methods and results:We analysed 243 adults (median age 62 years; 54% women) from the Sheba Healthspan Research Population (SHARP) study. Ultrasound-based CV age was estimated using focused cardiac POCUS with AI software. Blood age was calculated using the SenoClock platform from 45 routine biomarkers, and ECG age was derived using a convolutional neural network trained on >770 000 tracings. Correlations with chronological age and inter-clock agreement were examined. Participants were stratified into quintiles of US delta (US-chronological age). All three clocks correlated with chronological age (blood: r = 0.89, US: r = 0.74, ECG: r = 0.61; all P < 0.001). US-accelerated agers (top quintile) displayed a more adverse cardiometabolic profile, including higher diastolic blood pressure, body mass index, waist circumference, triglycerides, alongside lower HDL cholesterol, and more than double the prevalence of metabolic syndrome. Those with US age ≥2 years above chronological age had significantly higher odds of metabolic syndrome (odds ratio = 2.34, 95% confidence interval: 1.07-5.17, P = 0.034). Conclusion:AI-derived ultrasound-based cardiovascular biological age from handheld POCUS is associated with prevalent metabolic syndrome in this cross-sectional cohort, even when routine focused POCUS shows no abnormalities warranting referral.

Abstract Aims Hypertrophic cardiomyopathy (HCM) remains underdiagnosed due to limited access to expert imaging. We developed and validated a deep-learning (DL)-based echocardiographic model adaptable to point-of-care ultrasound (POCUS) for scalable HCM screening. Methods and results We retrospectively analysed 134 956 expert transthoracic echocardiograms (TTE) from 73 598 patients at Sheba Medical Center (2007–2022). A TTE-trained DL model integrating structural features and temporal motion patterns from parasternal long-axis and apical four-chamber views estimated HCM probability. Performance was evaluated in an independent test cohort and clinical subgroups. External validation used bedside POCUS studies from non-cardiologists with handheld devices. The test cohort included 12 096 patients with 119 confirmed HCM cases (prevalence 0.98%; median age 75 years, 57% male). HCM-positive patients showed increased expert TTE-measured septal (1.67 [1.5, 2.0] vs. 1.01 [0.9, 1.19] cm) and posterior wall thickness (1.1 [1.0, 1.3] vs. 0.9 [0.8, 1.0] cm) (P < 0.001). The model achieved excellent discrimination with an area under the curve of 0.982 (95% CI 0.966–0.993), sensitivity 88.2%, and specificity 97.3%, robust across subgroups. The POCUS cohort (n = 1047, median age 73 years, 55% male) represented multimorbid inpatients with 65 (6.2%) classified as screen-positive by the algorithm. These showed higher expert TTE-measured septal thickness (1.26 [1.07, 1.46] vs. 1.06 [0.9, 1.2] cm; 22% vs. 4% with IVS ≥1.5 cm; P ≤ 0.01). Among 49 (75%) POCUS-flagged positive patients with formal TTE and clinical data, 8 (16%) were confirmed by expert adjudication to have HCM. Specificity is limited by occasional confounding amyloidosis detection (4% of POCUS-flagged patients). Conclusion This DL-based model identifies HCM and demonstrates feasibility for POCUS screening, supporting earlier detection and broader diagnostic access.

Caregiving for cardiac patients is often accompanied by significant emotional and physical strain. While prior research has emphasized the contribution of support to mitigate caregivers' burden, limited attention has been paid to the support they may receive from the patients themselves. Attachment theory provides a crucial lens for understanding individual differences in how this specific kind of partner's support is perceived and utilized. Drawing on dyadic coping models and attachment theory, this study examines how support dynamics within the couple influence caregivers' burden, and whether these effects depend on caregivers' attachment orientations. Specifically, the study investigates whether caregivers' attachment anxiety and avoidance moderate the associations between (a) patient-reported support provided to the caregiver (b) caregiver perceived received support from the patient, and (c) levels of caregiver burden during cardiac rehabilitation. Eighty-eight heterosexual couples, in which the male partner had experienced a recent cardiac event, were assessed at the beginning and end of a three-month rehabilitation program. Measures included self-reported support (provided and received), attachment orientations, and multidimensional caregivers' burden. Overall support was not directly associated with reduced general burden. Caregivers high in avoidant attachment reported lower emotional burden when they perceived receiving greater support from their ill partner. No such effect was found among caregivers low in avoidance or those high in anxious attachment. These findings highlight the importance of considering both interpersonal and intrapersonal factors in understanding caregiver burden. Recognizing and addressing attachment-related patterns in caregiving dyads could inform targeted interventions aimed at reducing emotional distress and enhancing resilience in couples coping with cardiac illness.

Aims:Mitral and tricuspid regurgitation (MR and TR) are common in older adults and associated with substantial morbidity and mortality. While transthoracic echocardiography (TTE) is the diagnostic gold standard, access remains limited in many care settings. Artificial intelligence (AI)-based echocardiographic analysis may help address this diagnostic gap. Methods and results:We externally validated a deep learning algorithm developed by Aisap.ai using TTE studies from the Mayo Clinic Health System (2013-23). The model analyses echocardiographic images to classify atrioventricular regurgitation severity and was evaluated against cardiologist interpretations. Performance was assessed using binary (normal-mild vs. moderate-severe) and ordinal (normal, mild, moderate, severe) classification schemes. Among 1541 eligible TTEs, the model returned predictions for 578 studies (38%). Performance analysis was limited to these cases. The MR cohort included 280 studies and the TR cohort 298. For MR, the model achieved an area under the receiver operating characteristic curve (AUC) of 0.98 [95% confidence interval (CI): 0.97-0.99], with 91% accuracy, 95% sensitivity, and 89% specificity. For TR, the AUC was 0.96 (95% CI: 0.94-0.98), with 84% accuracy, 91% sensitivity, and 80% specificity. Conclusion:In cases where a prediction was generated, the model demonstrated high diagnostic performance in identifying clinically significant atrioventricular regurgitation. These findings support the feasibility of AI-assisted echocardiography in diverse populations, while underscoring the need for technical alignment between model requirements and local acquisition practices to ensure real-world applicability.

Physical activity plays a central role in cardiac rehabilitation, serving as a pivotal factor in recovery and secondary prevention. Interpersonal factors within patients’ romantic relationships play a critical role in the rehabilitation process, as partners tend to show behavioral concordance in their health behaviors. This study aimed to investigate the interdependence of physical activity between Acute Coronary Syndrome (ACS) patients and their partners, as well as the role of relationship satisfaction in moderating this concordance. Utilizing 21 daily diary assessments from couples, we examined co-fluctuations in patients’ and partners’ daily physical activity. A significant concordance in couples’ physical activity was found. However, relationship satisfaction did not moderate this association, suggesting that couples’ daily concordance occurred regardless of their average satisfaction level. These results highlight the potential influence of partners on each other’s health behaviors. Our findings suggest important implications for the development of dyadic interventions aimed at promoting physical activity as an integral component of cardiac rehabilitation and secondary prevention.

Chronic kidney disease (CKD) is a common comorbidity among patients with tricuspid regurgitation, yet its impact on tricuspid regurgitation outcomes is underexplored. This study examines how CKD affects the relationship between severe tricuspid regurgitation and overall survival. This is a retrospective cohort study of all adult patients (> 18 years old) evaluated at the Sheba Medical Center, between 2007 and 2022, who underwent transthoracic echocardiographic evaluation. It is based on the SHEBAHEART big data registry. Sheba Medical Center is the largest hospital in Israel with approximately 115,000 admissions per year. The echocardiographic reports together with the electronic medical records of all patients are the source for this study. Patients with missing creatinine data within one month of their echocardiography study, as well as those who underwent tricuspid regurgitation intervention, were excluded from the study. Patients were categorized into four groups, according to the presence and severity of tricuspid regurgitation and stratified by CKD stage. The primary outcome was all-cause mortality. The study included 78,147 patients (median age 67, IQR 55–78), with 2989 (4

BACKGROUND:Aortic valve replacement (AVR) is considered one of the most potent disease-modifying procedures among patients with severe aortic stenosis (sAS). Accordingly, we have witnessed a consistent increase in the procedure rates in recent years. Nevertheless, the elderly population, particularly octogenarians, remains relatively undertreated. The current study aims to document the disparities in AVR rates among octogenarians and its prognostic significance. METHODS:A vast database of Maccabi Health Services, the second largest health maintenance organisation in Israel, counting nearly 2.8 million members, was retrospectively analysed from 2005 to 2021 for all patients over 60 years, with a detailed echocardiography report compatible with a diagnosis of high-gradient sAS. The database was extracted using the MDClone healthcare data platform, generating synthetic data reliably representing the original population. All-cause mortality was set to be the primary outcome, and survival models using adjusted multivariable analyses for several clinical and echocardiographic parameters were applied. RESULTS:The cohort consisted of 1396 patients with high-gradient sAS (76±7 years) with 39% octogenarians. Octogenarians were less likely to undergo AVR (42% vs 60%, p<0.01) and presented more severe clinical profiles. AVR significantly reduced mortality and hospitalisations in both age groups, but octogenarians showed a pronounced survival benefit regardless of symptom status. A time-dependent analysis showed that AVR was associated with reduced all-cause mortality (HR 0.30, 95% CI 0.23 to 0.41, p<0.001) within the octogenarian group in 5 years. A similar protective effect was shown in the non-octogenarian group (HR 0.32, 95% CI 0.22 to 0.46, p<0.001). CONCLUSION:This study highlights significant treatment disparities in AVR among octogenarians with high-gradient sAS despite clear benefits in survival and reduced hospitalisations. The findings suggest the need for more inclusive treatment strategies, particularly for older patients, and underscore the importance of AVR in improving clinical outcomes in this population.

BACKGROUND:Chronic kidney disease (CKD) is a common comorbidity among patients with tricuspid regurgitation, yet its impact on tricuspid regurgitation outcomes is underexplored. This study examines how CKD affects the relationship between severe tricuspid regurgitation and overall survival. METHODS:This is a retrospective cohort study of all adult patients (> 18 years old) evaluated at the Sheba Medical Center, between 2007 and 2022, who underwent transthoracic echocardiographic evaluation. It is based on the SHEBAHEART big data registry. Sheba Medical Center is the largest hospital in Israel with approximately 115,000 admissions per year. The echocardiographic reports together with the electronic medical records of all patients are the source for this study. Patients with missing creatinine data within one month of their echocardiography study, as well as those who underwent tricuspid regurgitation intervention, were excluded from the study. Patients were categorized into four groups, according to the presence and severity of tricuspid regurgitation and stratified by CKD stage. The primary outcome was all-cause mortality. RESULTS:The study included 78,147 patients (median age 67, IQR 55-78), with 2989 (4%) having severe tricuspid regurgitation and 19,910 (25%) with an estimated glomerular filtration rate [eGFR] < 60 mL/min/1.73 m2. Over a median 4-year follow-up, 28,112 patients (36%) died. Both tricuspid regurgitation severity and CKD stage were associated with increased mortality risk (log-rank p < 0.001 for both). Adjusted models showed that compared to the none/trivial group, patients with mild, moderate, and severe tricuspid regurgitation had a 6%, 12%, and 35% higher risk of death, respectively (p < 0.001 for all). The association of tricuspid regurgitation with poor survival was CKD-dependent, with increased mortality risk of 56% vs. 23% among patients with eGFR < 60 vs. eGFR ≥ 60 (p for interaction < 0.001). The interaction analysis was no longer significant when right ventricular function was incorporated into the multivariable model. Subanalysis, limited to patients with isolated tricuspid regurgitation, yielded consistent results. CONCLUSIONS:The association between severe tricuspid regurgitation and poor survival is stronger in advanced CKD patients and may be modulated through right ventricular function.

With the rising prevalence and healthcare burden of structural heart disease and heart failure, artificial intelligence (AI) is playing an increasingly prominent role in echocardiography, with promising potential to enhance diagnostic accuracy and efficiency. We hypothesized that a Deep Neural Network (DNN) could be trained to identify and assess significant cardiac dysfunction using only a single apical 4-chamber echocardiographic view, rather than a comprehensive multi-view standard transthoracic echocardiogram. Echocardiographic reports and corresponding images from 121,767 unique patients, linked to clinical data (2007–2022), were analyzed by a designated DNN. A validation cohort, consisting solely of apical 4-chamber view clips, was used to assess the presence of significant (mild to moderate up to severe) mitral regurgitation (MR), tricuspid regurgitation (TR), right ventricular dysfunction (RVD), and assessment of left ventricular ejection fraction (LVEF). Additionally, a second cohort of 209 point-of-care focused cardiac ultrasound (FoCUS) examinations, performed by non-cardiologists, was analyzed by the DNN. For mitral regurgitation, the model achieved an AUC of 0.883, with a sensitivity of 0.744 and specificity of 0.844 (N = 20,313). For tricuspid regurgitation, it yielded an AUC of 0.913, with a sensitivity of 0.766 and specificity of 0.887 (N = 21,742). In assessing right ventricular dysfunction, the DNN reached an AUC of 0.942, with a sensitivity of 0.757 and specificity of 0.939 (N = 6,010). For LVEF assessment, the root mean square error (RMSE) was 4.78, and the mean absolute error (MAE) was 3.34 (N = 23,350). AI-driven analysis of a single apical 4-chamber echocardiographic view can provide valid estimates of significant valvular disease, right ventricular dysfunction, and LVEF with high sensitivity and specificity. These findings suggest that AI-based approaches could enable more efficient and widespread cardiac assessment.

Abstract Background Within cardiovascular medicine, AI-driven electrocardiogram (ECG) analysis has emerged as a powerful tool, capable of predicting clinically significant abnormalities. The increasing growth of wearable technologies for ECG monitoring underscores the importance of advancing AI methodologies. While Convolutional Neural Networks (CNNs) have traditionally been the go-to choice for ECG analysis, the rise of Vision Transformers (ViTs), a novel computer vision paradigm, raises the question of their potential superiority. Purpose To compare the performance of CNNs and ViTs in ECG analysis, and to assess the efficiency of lean models using truncated ECG recordings. Methods We trained a model using 12-lead ECGs as input for the binary classification of two outputs: Sex (Male or Female) and Left ventricular dysfunction (LVD, defined as below 35%), determined by echocardiography within 2 weeks of the ECG. Models were developed using full standard ECGs, single lead and truncated recordings (1,2 and 4 10 seconds). We compared CNNs (PyTorch) and ViTs (HuggingFace) performance. We analyzed the explainability of our findings, representing the AI model's focus of interest on ECGs from normal cases versus those with reduced ejection fraction, thereby validating its diagnostic discernment (Fig 1) Results We identified 150,691 and 29,422 patients with valid ECG tests for the Sex and LVD prediction models, respectively. For the Sex outcome, the AUROCs were 0.911 (95% CI: 0.908-0.914) and 0.898 (95% CI: 0.894-0.901) for the CNN-based model and the ViT-based models, respectively, with the difference statistically significant (P-Value < 0.001). For the LVD outcome, the AUROCs were comparable at 0.878 (95% CI: 0.864-0.892) and 0.866 (95% CI: 0.853-0.879) for the CNN-based model and the ViT-based models, respectively, (P-Value = 0.056). Furthermore, we found that 98.8% for the maximal predictive power of the models was achieved within a 2 second time frame of a 12 lead ECG. Similarly, a single lead based model achieved a relatively high AUC for the prediction of LVD. Conclusions AI-ECG models developed with ViT architecture did not surpass CNN-based models in sex classification and LVD identification. Our study highlights the significance of lean models allowing for rapid prediction within 2-second tracings or using a single lead, demonstrates the potential for streamlined AI-ECG applications in clinical practice.

This study examines the combined use of machine learning (ML) and expert judgment in predicting 30-day mortality for congestive heart failure (CHF) patients. It compares models using either expert-selected, ML-selected, or integrated features. The integrated model, merging expert and ML insights, outperforms others in predicting mortality risk, underscoring the value of combining human expertise and ML in clinical decision-making.

BACKGROUND:Age and sex can be estimated using artificial intelligence on the basis of various sources. The aims of this study were to test whether convolutional neural networks could be trained to estimate age and predict sex using standard transthoracic echocardiography and to evaluate the prognostic implications. METHODS:The algorithm was trained on 76,342 patients, validated in 22,825 patients, and tested in 20,960 patients. It was then externally validated using data from a different hospital (n = 556). Finally, a prospective cohort of handheld point-of-care ultrasound devices (n = 319; ClinicalTrials.gov identifier NCT05455541) was used to confirm the findings. A multivariate Cox regression model was used to investigate the association between age estimation and chronologic age with overall survival. RESULTS:The mean absolute error in age estimation was 4.9 years, with a Pearson correlation coefficient of 0.922. The probabilistic value of sex had an overall accuracy of 96.1% and an area under the curve of 0.993. External validation and prospective study cohorts yielded consistent results. Finally, survival analysis demonstrated that age prediction ≥5 years vs chronologic age was associated with an independent 34% increased risk for death during follow-up (P < .001). CONCLUSIONS:Applying artificial intelligence to standard transthoracic echocardiography allows the prediction of sex and the estimation of age. Machine-based estimation is an independent predictor of overall survival and, with further evaluation, can be used for risk stratification and estimation of biological age.