European Heart Journal - Digital Health

BackgroundDelayed or missed diagnosis of congenital heart disease (CHD) contributes to excess pediatric mortality worldwide. Echocardiography (echo) is central to diagnosing and triaging CHD, yet expert interpretation remains a scarce and maldistributed global resource. Artificial intelligence (AI) offers the potential to democratize diagnostics and extend expert-level interpretation beyond large academic centers, but its application in CHD remains underexplored. MethodsWe developed EchoFocus-CHD, an AI-enabled model for automated detection of 12 critical and 8 non-critical CHD lesions, individually and as composites. The composite critical CHD outcome was the primary endpoint. The model expands on a multi-task, view-agnostic architecture (PanEcho) with a transformer encoder to improve focus on relevant echo views. The model was trained (80%) and tested (20%) on the first echo per patient from Boston Childrens Hospital (BCH), with external validation on US and international studies from patients referred to BCH. ResultsThe internal and external cohorts included 3.4 million videos from 54,727 echos (median age at echo 7.1 [IQR, 0.2-15.0] years; 5.8% critical CHD; 23.6% non-critical CHD) and 167,484 videos from 3,356 echos (median age at echo 2.5 [IQR, 0.3-9.4] years; 29.4% critical CHD; 45.6% non-critical CHD), respectively. EchoFocus-CHD showed excellent internal ability to detect the composite critical CHD outcome (AUROC 0.94, LR+ 7.50, LR- 0.14) and individual critical lesions (AUROC 0.83-1.00), as well as composite non-critical CHD (AUROC 0.90, LR+ 5.00, LR- 0.23) and individual non-critical lesions (AUROC 0.70-0.96). Performance declined during external validation to detect critical CHD (AUROC 0.77), coinciding with greater expert disagreement on external cases ({kappa}=0.72 versus 0.82 for internal cases). Explainability analyses demonstrated that the model prioritized the same clinically relevant views (parasternal long-axis, parasternal short-axis, and subxiphoid long-axis) across internal and external cohorts, while UMAP analysis revealed a domain shift between cohorts. Retraining on all available US patients attenuated domain shift, improving international critical CHD detection (AUROC 0.87) and calibration. ConclusionsEchoFocus-CHD shows promise for automated CHD detection and highlights the need to address domain shift for real-world deployment. By identifying high-risk CHD lesions, this approach could support triage, prioritize expert review, and optimize resource allocation, advancing more equitable global cardiovascular care.

BackgroundAccurate assessment of left ventricular outflow tract (LVOT) gradients is critical for hypertrophic cardiomyopathy (HCM) management, yet Doppler-based measurements are technically demanding and require expertise. ObjectiveTo develop a multi-view deep learning model capable of classifying LVOT obstruction (> 20mmHg) using routine 2D echocardiographic windows without reliance on Doppler imaging. MethodsWe trained and externally validated a cross-attention-based video-to-video fusion framework that integrated EchoPrime-derived video representations from three standard transthoracic echocardiographic views to classify LVOT gradients. ResultsTraining was performed on a derivation cohort (N = 1833) from a tertiary care system in the United States, with model performance evaluated on an internal held-out test set (N = 275) and a Korean external validation cohort (N = 46). Single-view baselines showed limited discrimination (external AUROCs 0.47-0.70). Conversely, domain-specific foundational model (EchoPrime) achieved superior single-view performance (AUROCs 0.75-0.80 internal; 0.79-0.83 external), highlighting the importance of echo-specific pretraining and temporal modeling. The proposed multi-view fusion further enhanced predictive performance, with the late fusion model reaching an AUROC of 0.84 on the external cohort with significant population-shift. ConclusionsThese results suggest LVOT physiology is encoded in routine 2D imaging and can be leveraged for clinically relevant gradient classification without Doppler input- proposed AI-guided strategy demonstrates substantial cost savings compared with the screen-all approach. By integrating complementary spatial-temporal information across multiple views, our approach generalizes robustly across populations and may enable real-time decision support, extend LVOT assessment to portable or resource-limited settings, and complement Doppler-based evaluation for longitudinal HCM management.

BackgroundAcute heart failure (AHF) exhibits marked heterogeneity in diastolic hemodynamics, yet comprehensive echocardiographic assessment of diastolic function (DF) and filling pressure (FP) is often infeasible. We evaluated whether artificial intelligence-enabled electrocardiography (AI-ECG) could provide scalable DF grading and FP estimation in hospitalized AHF patients. MethodsWe retrospectively studied adults hospitalized for AHF across Mayo Clinic sites (2013-2023) who received [≥]1 dose of intravenous loop diuretic and had paired 12-lead ECG and TTE. The previously validated AI-ECG DF model was applied without retraining to generate four DF grades and a continuous FP probability. Clinical outcomes were all-cause mortality and heart failure rehospitalization. Associations with clinical severity markers and echocardiographic indices were examined. Kaplan-Meier survival analysis and adjusted multivariable Cox proportional hazards models were performed. Exploratory analyses examine the kinetics of change in FP probability and impact on mortality. ResultsAmong 11,513 patients (median age 75 years, 39% female), AI-ECG DF grading was feasible in 100%, whereas echocardiographic DF was indeterminate in 44% of clinically eligible patients. In 2,582 patients with determinate echocardiographic DF, AI-ECG FP probability discriminated TTE Grade 2-3 dysfunction with AUC 0.85 (95% CI 0.83 - 0.86). Higher AI-ECG DF grades were associated with higher comorbidity burden, worse NYHA class, elevated NT-proBNP, higher MAGGIC scores, elevated PCWP, and more advanced structural remodeling. After multivariable adjustment, AI-ECG DF remained independently associated with mortality (hazard ratio [HR] 1.25, 95% CI 1.16-1.35 for Grade 2; HR 1.44, 95% CI 1.33-1.56 for Grade 3 versus Normal/Grade 1). Combining AI-ECG DF with MAGGIC scores yielded ordered risk gradients, with highest mortality in patients with both high MAGGIC and Grade 2-3 DF. Among patients with serial ECGs, improvement in FP probability was independently associated with lower mortality (HR 0.85, 95% CI 0.79-0.91), whereas worsening did not show a consistent adverse gradient beyond baseline DF. ConclusionsIn a large, geographically diverse AHF cohort, AI-ECG DF grading was universally feasible, correlated with established hemodynamic severity markers, and provided independent prognostic information beyond established risk factors, supporting its role as a pragmatic, scalable diastolic biomarker in AHF. CLINICAL PERSPECTIVEO_ST_ABSWhat Is New?C_ST_ABSO_LIIn 11,513 hospitalized acute heart failure (HF) patients, artificial intelligence-enabled electrocardiography provided diastolic function grading in 100% of patients from a single 12-lead ECG without requiring additional clinical variables, compared with 56% feasibility for guideline-based echocardiography grading. C_LIO_LIAI-ECG diastolic function grades correlated with established marker of severity (NYHA functional class, NT-proBNP, MAGGIC risk scores, and pulmonary capillary wedge pressure) and remained independently associated with both mortality and HF rehospitalization after multivariable adjustment. C_LIO_LISerial AI-ECG measurements identified post-discharge filling pressure trajectories, with improvement independently associated with 15% lower mortality, a first demonstration that longitudinal ECG assessment can track post-discharge hemodynamic recovery. C_LI What Are the Clinical Implications?O_LIAI-ECG transforms the universally obtained 12-lead ECG into an actionable hemodynamic biomarker that addresses the critical gap when echocardiographic diastolic function assessment is indeterminate or unavailable in acute HF patients. C_LIO_LIDespite markedly different hemodynamic severity and long-term outcomes across AI-ECG diastolic function grades, hospitalization length of stay did not differ, suggesting advanced diastolic dysfunction represents occult risk not easily recognized during routine acute care and highlighting the need for improved post-discharge risk stratification. C_LIO_LIThe continuous filling pressure probability metric enables longitudinal monitoring of post-discharge hemodynamic status using serial routine ECGs, potentially identifying patients requiring intensified follow-up or specialist referral. C_LI

BackgroundEchocardiography (echo) is a cornerstone of pediatric cardiology, yet access to expert interpreters is limited worldwide, particularly in low-resource and rural settings. Artificial intelligence (AI) offers a mechanism to broadly deliver expert-level precision and standardize measurements, yet AI for comprehensive automated measurements in pediatric and congenital heart disease (CHD) echo remains underdeveloped. MethodsWe created EchoFocus-Measure, an AI platform that automatically extracts 18 quantitative parameters and 10 qualitative assessments from full echo studies. The method extends a multi-task, view-agnostic architecture (PanEcho) with a study-level transformer to prioritize diagnostically informative views. Training (80%) and internal testing (20%) were performed on echos from Boston Childrens Hospital (BCH), with external evaluation on outside referral studies. Left ventricular ejection fraction (LVEF) was the primary endpoint. ResultsThe internal cohort included 11.4 million videos from 217,435 echos (60,269 patients; median age 8.5 years; median LVEF 61%), and external validation encompassed 289,613 videos from 3,096 echos (2,506 patients; median age 3.5 years; median LVEF 62%). For LVEF, EchoFocus-Measure exhibited a median absolute error (MAE) of 2.8% internally and 3.8% externally, maintaining accuracy across infants (MAE 3.2%) and complex CHD lesions (e.g., MAE 4.0% for L-loop transposition of the great arteries). EchoFocus-Measure improved upon the PanEcho benchmark (MAE 7.5% for infants; 13.1% for L-loop transposition). Discrepant case (>50th percentile error) adjudication of LVEF demonstrated that model errors (MAE 2.4%) were within human variability (MAE 3.7%). For qualitative measures, EchoFocus-Measure performed well internally (AUROC 0.88-0.95) and modestly externally (AUROC 0.73-0.86). Explainability analyses highlighted model focus on clinically appropriate echo views for LVEF estimation (apical four-chamber, parasternal short/long) and mitral regurgitation assessment (apical four-chamber color Doppler, parasternal short/long color Doppler). ConclusionsEchoFocus-Measure delivers rapid and reliable automated echo measurements across ages and lesions within diverse internal and real-world external cohorts, serving as a step toward scalable, global access to high-quality pediatric cardiovascular care.

Background and AimsAccurate classification of mitral stenosis (MS) remains a significant clinical challenge. This study aimed to develop an artificial intelligence (AI) framework to automatically detect clinically significant MS from echocardiography. MethodsWe developed EchoNet-MS, an open-source end-to-end integrated approach combining video based convolutional neural networks to assess MS severity and differentiate rheumatic etiology from echocardiography and validated its performance across four cohorts. ResultsEchoNet-MS was trained and validated in total of 431,612 videos from 44,671 studies from three different healthcare system. Combining assessments from multiple echocardiographic videos, the model was trained on a Kaiser Permanente Northern California (KPNC) cohort of 8,677 studies from 7,576 patients with a range of MS severity. The model was validated on a KPNC held-out test cohort (N=1,623) and a temporally distinct cohort (N=19,206), as well as Stanford Healthcare (SHC) cohort (N=3,333) and Cedars-Sinai Medical Center (CSMC) cohort (N=72,909). EchoNet-MS achieved excellent discrimination of severe MS with AUC 0.937 [95% CI: 0.913 - 0.958] in the KPNC held-out cohort, 0.994 [0.986 - 0.999] in the temporally distinct cohort, 0.991 [0.986 - 0.995] in SHC, and 0.973 [0.958 - 0.987] in CSMC. The model achieved excellent performance in classifying both rheumatic or non-rheumatic MS with AUC ranging from 0.890 and 0.967. ConclusionsEchoNet-MS accurately assesses MS severity and etiology using information from multiple echocardiographic views. Its strong performance generalizes robustly to external cohorts and shows potential as an automated clinical decision support tool.

Echocardiography is the most widely used cardiac imaging modality, yet artificial intelligence-enabled interpretation remains limited by the inability of existing models to integrate visual assessment, quantitative measurement, and clinical reasoning within a unified framework. Here we present EchoAtlas, the first autoregressive vision-language model developed for echocardiographic interpretation. Trained on over 12.9 million question-answer pairs derived from approximately 2 million echocardiogram videos, EchoAtlas achieves 0.966 accuracy on multiple-choice questions in our internal test set and establishes a new state-of-the-art on the public MIMIC-EchoQA benchmark (0.699 vs. 0.508 previously). EchoAtlas also provides accurate quantitative measurements, segment-level regional wall motion assessment, longitudinal comparison, and diagnostic reasoning across diverse question formats -- capabilities not previously demonstrated in this domain. These results highlight the potential of autoregressive vision-language models as a foundation for interactive echocardiographic interpretation, representing an early step toward scalable, auditable artificial intelligence systems in cardiology practice.

Background Clinical prediction models degrade when deployed across hospitals, yet retraining requires technical expertise, labeled data, and regulatory re-approval. We investigated whether post-hoc retrieval augmentation of a frozen model's output, analogous to retrieval-augmented methods in natural language processing, can mitigate this degradation without any parameter modification. Methods We developed the Post-hoc Retrieval Augmentation Module (PRAM), which combines predictions from a frozen base model with outcome information retrieved from similar patients in a local patient bank. Five base models (logistic regression through CatBoost) and three retrieval strategies were evaluated on 116,010 ICU patients across three databases (MIMIC-IV, MIMIC-III, eICU-CRD) for acute kidney injury (AKI) and mortality prediction. A bank size deployment simulation modeled performance from zero to full local data accumulation, complemented by source bank cold start, stress tests, and calibration experiments. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC). Results Retrieval benefit was inversely associated with base model complexity ({rho} = -0.90 for AKI, -1.00 for mortality): simpler models benefited more, consistent with retrieval capturing residual signal unexploited by the base model. PRAM showed a statistically significant monotone dose-response between bank size and prediction performance across all six outcome-target combinations (Kendall {tau} trend test, q = 0.031 for all). At the pre-specified primary comparison (bank = 5,000), the improvement was confirmed for the two largest-shift settings (eICU-CRD AKI: {Delta}AUROC = +0.012, q < 0.001; eICU-CRD mortality: {Delta}AUROC = +0.026, q < 0.001). Pre-loading a source bank bridged the cold-start gap, providing an immediate performance gain equivalent to approximately 2,000 to 5,000 local patients. Conclusions PRAM provides a parameter-free adaptation mechanism that requires no model retraining, gradient computation, or regulatory re-evaluation at the deployment site. Effect sizes were modest and did not reach cross-model superiority, but the consistent dose-response pattern and the absence of retraining requirements establish retrieval-based adaptation as a viable approach for clinical model transportability. The retrieval mechanism additionally opens a pathway toward case-based interpretability, where predictions are accompanied by identifiable similar patients from the deploying institution.

Background: Transthyretin cardiomyopathy (ATTR-CM) is a progressive, potentially fatal disease requiring accurate risk stratification. Echocardiography is the first-line imaging modality, with AI-based tools increasingly applied for automated analysis, yet their prognostic value remains unknown. Objectives: To examine the prognostic value of AI-derived echocardiographic measurements and their incremental value beyond biomarker staging in ATTR-CM. Methods: This retrospective study included patients from two ATTR-CM registries. Baseline echocardiograms were analyzed using the fully automated AI-based software Us2.ai. Prognostic performance was assessed by Kaplan-Meier analysis, Cox regression, and ROC curves. A two-parameter echocardiographic staging system combining left ventricular (LV) global longitudinal strain (GLS) and right ventricular (RV) fractional area change (FAC) stratified patients into low (both normal), intermediate (one abnormal), and high risk (both abnormal). Results: Among 347 patients (91% male, median age 78 years), 141 experienced all-cause death or heart failure hospitalization over a median follow-up of 2.4 years. In multivariable analysis, AI-derived LV-GLS (HR 1.13 [1.03-1.25], p=0.011) and RV FAC (HR 0.96 [0.93-0.99], p=0.014) were independent outcome predictors. Echo staging stratified risk into groups with 3-fold (95% CI 1.70-5.91) and 6-fold (95% CI 3.22-10.30) increased hazard compared to low risk (p<0.001), with incremental prognostic value beyond National Amyloidosis Centre (NAC) staging and age (chi-square from 53 to 80; p<0.001). AI and human measurements showed comparable 1-year predictive performance (all p>0.05). Conclusion: AI-derived echocardiographic measurements demonstrate independent and incremental prognostic value beyond biomarker-based NAC staging in ATTR-CM, comparable to human measurements, supporting their integration into clinical risk stratification.

ObjectivesTo identify unique echocardiographic signatures associated with TTR+ carrier status preceding onset of cardiac amyloidosis. BackgroundCarrier status for the most common pathogenic TTR variant in the United States, Val142Ile (V142I), found in 4% of African Americans (AA) and 1% of Hispanic/Latino (H/L) individuals, confers a 40-60% lifetime risk of developing variant transthyretin amyloidosis (ATTRv), including cardiac amyloidosis (CA) and heart failure (HF). Myocardial amyloid deposition is believed to progress over many years. Genomic screening programs and familial cascade genetic testing are increasingly uncovering pre-symptomatic TTR+ carriers, yet no guidelines exist to pragmatically risk stratify these individuals for CA. MethodsV142I+ carriers (cases) without prior diagnoses of amyloidosis or HF were identified among BioMe biobank participants with available exome sequencing data linked to electronic health records (EHRs) including at least one available echocardiogram. Controls were biobank participants with normal TTR sequencing who were age-, sex- and ancestry-matched to cases. Speckle-tracking echocardiography (STE) was applied to images and conventional and strain measurements were evaluated by univariate analyses. A random forest model was trained using a minimal redundancy maximal relevance (mRMR, applied to mitigate overfitting) feature set and evaluated by 5-fold cross-validation to minimize optimism bias. Discriminatory performance was assessed using the area under the receiver operating characteristic curve (AUC). Results49 TTR+ (100% V142I, median age 61 years, 69.4% female) and 45 matched TTR-biobank participants were included in the model development cohort. STE generated approximately 200 features. Univariate analyses revealed no significant differences between carriers and controls on any individual strain or conventional echocardiographic measurements including global longitudinal, right ventricular and left atrial strain. mRMR feature selection resulted in a set of 15 features retained for all downstream modeling, integrating global amyloid signatures, regional inferolateral strain abnormalities, layer-specific deformation, and mechanical timing heterogeneity. Using this feature set, the model achieved good discrimination (AUC=0.76). Feature importance analysis highlighted relative apical sparing, inferolateral strain reduction, and basal-apical timing gradients as key contributors to model performance. External validation (n=115) confirmed good model discrimination (AUC=0.781, 95% CI: 0.688-0.869, sensitivity 0.983). ConclusionsMachine learning applied to routinely acquired echocardiographic data can identify subtle myocardial abnormalities associated with TTR V142I carrier status prior to development of CA. Key model features are physiologically relevant to known echocardiographic characteristics of overt CA. Genotype-guided echocardiographic surveillance may be a scalable strategy for early detection of CA risk.

BackgroundVarious measurements around the aortic valve are typically made on computed tomography angiograms (CTAs) before transcathether aortic valve replacement (TAVR) for aortic stenosis (AS), but their collective prognostic inference on periprocedural conduction disturbances (CDs) is not known. Here, we aimed to use unsupervised machine learning (UML) to analyze a multitude of pre-TAVR CTA features and uncover patient subphenotypes with differential risks of CDs. MethodsTwelve nonredundant features involving the aortic valve, aortic root, and ascending aorta were extracted from the CTAs of 660 AS patients. UML of these features using agglomerative hierarchical clustering was performed on separate male and female datasets, with the optimal number of clusters determined by 30 cluster indices. Multivariable logistic regression was conducted to assess the dependence of CDs on cluster type and the latters incremental prognostic value over conventional risk factors. ResultsThree male clusters were optimally identified (M1-M3): M1 was associated with small valve leaflet calcification loads and aortic root dimensions; both M2 and M3 were associated with large valve leaflet calcification loads and a wide aortic root, but the aortic root was shorter in M2 than M3. Two female clusters were optimally determined (F1-F2): F2 was associated with larger valve leaflet calcification loads and aortic root dimensions. By logistic regression analysis, compared to M1 (reference), M2, but not M3, was more associated with CDs (ORM2/M1=2.15, P=0.032; ORM3/M1=2.12, P=0.085), with no difference between M3 and M2 (ORM3/M2=0.986, P=0.974) or between F1 and F2 (ORF2/F1=1.294, P=0.581). Including cluster type as a predictor in a regression model of CDs containing conventional risk factors as covariates improved the goodness-of-fit (P=0.020). ConclusionsUML of pre-TAVR CTAs can reveal subgroups of male patients with differential risks for CDs and improve prognostication over conventional risk factors. UML-augmented pre-TAVR CTAs may help better guide personalized strategies to minimize CDs.

Background and AimsHeart failure with preserved ejection fraction (HFpEF) exhibits profound phenotypic heterogeneity, which likely contributes to variable therapeutic response. We developed a physiology-informed digital twin-AI framework to predict individual hemodynamic and myocardial energetic responses to accelerated atrial pacing and tested whether simulated physiologic response corresponds to responders in the myPACE randomized clinical trial. MethodsPatient-specific digital twins were constructed for 146 HFpEF patients and used to train a variational autoencoder that generated a virtual HFpEF population (n = 25,000). The model simulated pacing-induced changes in left atrial pressure (LAP), systolic blood pressure (SBP), cardiac output (CO), and cardiac efficiency (CE; derived from myocardial oxygen-demand estimates). These simulations served as labels to train classifiers based on clinical variables available in myPACE, allowing us to examine associations with clinical end points and test a hypothesized relationship between CE and treatment response. ResultsSimulations revealed heterogeneous physiological responses, with 95.6% of virtual patients showing reduced LAP, 47.0% an SBP reduction greater than 8.5 mmHg, 93.8% increased CO, and 36.1% improved CE. Classifiers reproduced these patterns with high fidelity. In the myPACE trial, patients classified as having CE improvement or a larger SBP reduction experienced significantly greater 1-month improvements in quality-of-life scores and larger NT-proBNP reductions. ConclusionsA physiology-informed digital twin-AI framework can predict hemodynamic and energetic responses corresponding to clinical benefit in HFpEF patients receiving accelerated atrial pacing. CE improvement functioned as a mechanistic indicator, while SBP reduction served as an accessible clinical correlate, offering mechanistically grounded guidance for patient-specific pacing and motivating prospective validation. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=122 SRC="FIGDIR/small/26347199v1_ufig1.gif" ALT="Figure 1"> View larger version (63K): org.highwire.dtl.DTLVardef@4a550eorg.highwire.dtl.DTLVardef@163b85org.highwire.dtl.DTLVardef@19db16dorg.highwire.dtl.DTLVardef@1eb6cf5_HPS_FORMAT_FIGEXP M_FIG C_FIG

BackgroundArtificial intelligence (AI) has emerged as a promising tool for interpreting 12-lead electrocardiograms (ECGs), with the potential to enhance diagnostic accuracy for arrhythmia detection. However, published studies vary widely in methodology and validation strategy, warranting a quantitative synthesis of diagnostic performance. MethodsA systematic review and meta-analysis was conducted according to the PRISMA-DTA 2018 guidelines and registered in PROSPERO (CRD420251027264). Searches were performed in MEDLINE, Embase, and Cochrane Library through September 2025 without language restrictions. Studies evaluating AI algorithms for arrhythmia detection using 12-lead ECGs were included. Data on sensitivity, specificity, and area under the curve (AUC) were extracted. Pooled estimates were generated using a bivariate random-effects model. Risk of bias was assessed with QUADAS-2, and the certainty of evidence was quantified using GRADE. Results20 studies were included in the meta-analysis, encompassing over 5.5 million ECGs. The pooled sensitivity, specificity, and AUC for AI-based arrhythmia detection were 94.0% (95% CI 90.8-96.2; I{superscript 2} = 96.9%), 98.7% (95% CI 97.3-99.3; I{superscript 2} = 98.3%), and 0.982 (95% CI 0.965-0.986), respectively. Detection of atrial fibrillation (AF) yielded a sensitivity of 92.6% (95% CI 86.4-96), a specificity of 99.1% (95% CI 98.4-99.5), and an AUC of 0.988. Convolutional neural networks (CNN) specifically demonstrated a sensitivity of 97.6%, specificity of 98.7%, and an AUC of 0.982 for overall arrhythmia detection. When limited to external validation (n=6), the sensitivity was 96.9% (95% CI 89.2-99.1), specificity was 95.6% (95% CI 77.6-99.3), and AUC was 0.983. No significant publication bias was detected, and the overall certainty of evidence was rated as high. ConclusionsAI models applied to 12-lead ECGs demonstrate excellent diagnostic performance for arrhythmia detection. Findings support potential integration into clinical workflows, particularly in settings with limited cardiology expertise. Given substantial heterogeneity, standardized datasets and multicenter prospective validation are essential to ensure effective and equitable implementation. What is KnownO_LIArtificial intelligence has been increasingly applied to 12-lead electrocardiograms for arrhythmia detection, with multiple studies reporting high diagnostic accuracy. C_LI What the Study AddsO_LIThis meta-analysis demonstrates consistently high diagnostic performance of artificial intelligence for arrhythmia detection on 12-lead ECGs, including atrial fibrillation and externally validated models. C_LIO_LIThe substantial heterogeneity observed underscores the need for standardized datasets and multicenter prospective validation before widespread clinical implementation. C_LI

BACKGROUNDConventional electrocardiography (ECG) has limited diagnostic accuracy for detecting coronary artery disease (CAD) in patients with stable chest pain. Advanced electrocardiography (A-ECG) may improve diagnostic performance. The study aimed to derive, externally validate, and prognostically validate an explainable A-ECG score for detecting CAD on coronary computed tomography angiography (CCTA). METHODSParticipants attending an outpatient rapid access chest pain clinic (RACC) underwent a standard 12-lead ECG and CCTA. Any CAD was defined as any calcified or non-calcified plaque. Elastic net with nested resampling was used to derive an A-ECG score using measures from the conventional ECG, derived vectorcardiography, and measures of waveform complexity. RESULTSIn the derivation cohort (n=171, age 59{+/-}13 years, 60% male), n=99 (58%) had any CAD on CCTA. A seven parameter A-ECG score to detect any CAD was derived. In an external validation cohort (n=773, age 57{+/-}12 years, 49% male, n=433 (56%) with any CAD), the score had an area under the receiver operating characteristic curve [95% confidence interval] of 0.66 [0.63-0.70] for detecting any CAD, and 0.72 [0.68-0.76] for detecting any coronary artery calcification. In the UK Biobank (n=27,239, 966 events, follow-up 1.9 [0.7-4.4] years, age 66{+/-}8 years, 50% female), higher A-ECG scores were associated with cardiovascular events even after adjusting for age, sex and cardiovascular risk factors (p<0.001). CONCLUSIONSAn explainable A-ECG model, incorporating demographic and electrocardiographic features, demonstrated modest but externally reproducible discrimination for CCTA-defined coronary atherosclerosis and independent prognostic association in a large population cohort. This scalable, low-cost approach may aid triage and risk stratification in chest pain pathways.

BackgroundTraditional heart transplant registries often lack the granularity required for deep phenotyping and rely on labor-intensive manual abstraction. We describe the methodology and validation of a next-generation, automated, multi-source registry designed to address these limitations. MethodsUtilizing a High-Performance Computing environment, we integrated structured data from Epic data warehouses (Clarity and Caboodle), external molecular diagnostics, and verified UNOS survival records. A custom deterministic rule-based Natural Language Processing (NLP) engine was developed to extract echocardiographic measures, rejection grades, and vasculopathy scores from over 21,000 unstructured clinical reports. ResultsThe Houston Methodist J.C. Walter Jr. Transplant Center Precision Registry and Platform-Heart (TCPR-Heart) captures 1,687 heart transplants (1,636 patients) spanning the years 1984-2025. The TCPR-Heart comprises 1,054 transplants with active clinical follow-up: 555 transplants were extracted and abstracted from our modern electronic health record (EHR) in the decade since deployment, providing access to data throughout the patients course of heart transplant; 427 were legacy active transplants (transplanted pre-2016 with continued follow-up), and 72 were external transplants (transplanted elsewhere but followed at Methodist). Additionally, the registry houses a historic cohort of 633 transplants (last follow-up < June 2016) with limited variables. Automated deep phenotyping successfully generated longitudinal data trends across clinical domains, including immunosuppression strategies, rejection, immunologic HLA data, renal function, metabolic profiles, vasculopathy, graft function, hospitalization burden and survival information. ConclusionThis automated framework unifies clinical, administrative, and molecular data streams. By leveraging an automated, regularly updated registry, we established a scalable, high-fidelity data source as a foundation for further innovations and novel applications based on an expertly curated and validated data source.

Backgound: Accurate assessment of diastolic function and left ventricular (LV) filling pressure is central to heart failure diagnosis and risk stratification. Contemporary guideline algorithms rely on complex parameters that are not consistently available in routine clinical practice. Objective: To compare the diagnostic and prognostic performance of the 2016 American Society of Echocardiography/European Association of Cardiovascular Imaging (ASE/EACVI) and 2025 ASE guidelines with a deep learning model based on routinely acquired echocardiographic variables. Methods: This study evaluated the guideline-based algorithms and a deep learning model in participants from the Atherosclerosis Risk in Communities (ARIC) cohort (n=5450) for prognostication and two invasive hemodynamic validation cohorts from the United States (n=83) and Japan (n=130) for detection of elevated left ventricular filling pressure. Results: In the ARIC cohort, the deep learning model demonstrated superior prognostic performance compared with the 2016 and 2025 guidelines (C-index: 0.676 vs. 0.638 and 0.602, respectively; both p<0.001). Similar findings were observed among participants with preserved ejection fraction (C-index: 0.660 vs. 0.628 and 0.590; both p<0.001), with improved performance compared with the H2FPEF score (C-index: 0.660 vs. 0.607; p<0.001). In the US hemodynamic validation cohort, the deep learning model showed higher diagnostic performance than the 2025 guidelines (AUC: 0.879 vs. 0.822; p=0.041) and similar performance compared with the 2016 guidelines (AUC: 0.879 vs. 0.812; p=0.138). In the Japanese hemodynamic validation cohort, the deep learning model outperformed both guidelines (AUC: 0.816 vs. 0.634 and 0.694; both p<0.05). Conclusions: A deep learning model leveraging routinely available echocardiographic parameters demonstrated improved diagnostic and prognostic performance compared with contemporary guideline-based approaches, potentially offering a scalable alternative for assessing diastolic function and left ventricular filling pressures.

Background: Current risk assessment tools for guiding direct oral anticoagulant (DOAC) therapy for patients with atrial fibrillation (AF) based on clinical risk factors demonstrate modest predictive performance limiting clinical impact. Additionally, while guidelines recommend periodic reassessment of risk over time, there remains an absence of modelling solutions for capturing evolving risk in AF patients. Methods: Using UK electronic health records, we developed and validated the Transformer-based Risk assessment survival model (TRisk), an artificial intelligence model that predicts 12-month thromboembolic and bleeding events in AF patients by leveraging temporal patient journeys up to baseline. A cohort of 411,850 prevalent non-valvular AF patients aged [&ge;]18 years between 2010 and 2020 was identified from 1,442 English general practices. Practices were randomly allocated to derivation (n=1,079) and external validation (n=363) cohorts. TRisk was compared with CHA2DS2-VASc and CHA2DS2-VA for thromboembolic event prediction, and HAS-BLED and ORBIT for bleeding prediction, with subgroup analyses by sex, age, and baseline characteristics. A second validation of TRisk was also performed on 16,218 US AF patients between 2010 and 2023. A decision model compared outcomes and healthcare costs for TRisk versus standard care. Findings: TRisk achieved higher discrimination for thromboembolic event prediction (C-index: 0.82; 95% confidence interval [CI]: [0.81, 0.83]) as compared to CHA2DS2-VASc (0.71 [0.70, 0.73]) in UK validation. Application of TRisk to US data yielded similar C-index: 0.82 (0.80, 0.84). For bleeding prediction, TRisk (C-index: 0.70 [0.69-0.71]) outperformed both HAS-BLED (0.63; [0.61, 0.64]) and ORBIT (0.64; [0.63, 0.65]), with comparable US results (0.71; [0.69, 0.74]). The model remained well-calibrated across both populations and performed equitably across subgroups, including by race and during the COVID-19 pandemic. Impact analyses showed TRisk could reduce DOAC prescriptions by 8% in the UK and 7% in the US relative to guideline-recommended approaches, while preventing at least as many thromboembolic events. This refined approach would generate annual healthcare savings of GBP 5.5 million and USD 456.2 million in the UK and US respectively among patients initiating DOACs, rising to GBP 48.6 million and USD 1.8 billion when extended to all AF patients on DOACs. Interpretation: TRisk enabled more precise prediction for both thromboembolic and bleeding events across AF populations in UK and US compared to established clinical scoring systems. Incorporating TRisk into routine AF care would result in substantial cost savings without compromising the identification of true high-risk patients. Funding: None

BackgroundMany patients with heart failure (HF) remain undiagnosed until acute hospital admission for decompensation. Think-HF is a clinical decision support tool (CDST) designed to identify possible undiagnosed HF in primary care and trigger timely, guideline recommended assessment. MethodsThink-HF was developed through epidemiological evidence and codesign. Integrated within SystmOne, it analyses routine primary care data (codes, medications, test results, free text) to detect missed or emerging HF signals. When a record is opened, if a patient has [&ge;] two comorbidity indicators plus an HF-suggestive symptom, an alert is triggered and a structured template opens with one-click options for natriuretic peptide (NP) testing, echocardiography, specialist referral and coding. Additional algorithms identify unresolved investigations, coding inconsistencies and medication-based signals, generating a marker on the patients record. A mixed-methods feasibility study across six primary care practices assessed reach, usability, acceptability, and early implementation signals, using the RE-AIM framework. ResultsSix socioeconomically and ethnically diverse general practices participated (63 GPs; 78,640 patients [47.5% women, 59.4% White, 17.8% South Asian, 8.5% Black, 3% Chinese and 2.8% mixed ethnicities]). At baseline, 876 patients (1.1%) met the main alert criteria and 2,805 (3.6%) met additional algorithm criteria. Of 801 patients on the HF register, 665 (83%) lacked a refined HF left ventricular phenotype code. During four months of testing, Think-HF generated 299 clinically relevant main trigger alerts (75% during consultations). Early improvements included (i) 31 new HF diagnoses (+4.2%), with higher gains (+10%) in practices with lower baseline prevalence; (ii) a 14% increase in refined HF phenotype coding; (iii) fewer raised NT-proBNP results without follow-up (-7%); and (iv) fewer patients prescribed loop diuretics without NP testing (-14%, up to -45% in one practice with pharmacist-supported review). Clinicians reported improved awareness, more systematic assessment, and better follow-up of missed investigations or coding anomalies. Identified gaps aligned with patient-reported delays and misattributed symptoms. ConclusionsThink-HF is feasible, acceptable and well aligned with routine primary care workflows. Early gains in diagnostic processes and coding accuracy highlight its potential to improve patient care. Team-based implementation will be essential for scale-up. Larger evaluation is required to assess clinical impact.

The recently published EHRA/EACVI consensus statement on a standardized bi-atrial regionalization provides new opportunities for consistent regional analyses across patients, imaging modalities and clinical centers. To make this standardized regionalization widely accessible, we developed the open-source software DIVAID, which automatically divides bi-atrial geometries according to the proposed regions, ensuring consistency, reproducibility and operator independence. We evaluated the accuracy of the algorithm by comparing its results to manual expert annotations across 140 geometries from multiple modalities and centers. Veins were automatically clipped correctly in 81% and orifices annotated correctly in 100% of cases. The median (interquartile range; IQR) Dice similarity coefficient (DSC) for left atrial regions was 0.98 (0.96-1.00) for DIVAID-expert and 0.98 (0.94-1.00) for inter-expert comparisons. For right atrial geometries, DSC was higher for DIVAID-expert than for inter-expert comparisons at 0.90 (0.80-0.95) and 0.88 (0.74-0.94), respectively. To assess the accuracy of regional boundaries, we computed the mean average surface distance (MASD) for boundaries derived from automatic or manual annotations. The median (IQR) MASD between DIVAID and experts was 0.17 mm (0.03-0.78) and 1.93 mm (0.65-3.96) in the left and right atrium, respectively. To conclude, DIVAID robustly divides anatomically diverse bi-atrial geometries according to the 15-segment model, while outperforming cardiac experts in both speed and consistency, and demonstrating an accuracy of regional boundaries comparable to the spatial resolution of cardiac imaging modalities. By providing automated, consistent atrial regionalization, DIVAID enables large-scale, standardized regional analyses and data-driven investigation of harmonized, multi-dimensional datasets, which may advance atrial arrhythmia research and personalized treatment strategies.

BackgroundCardiovascular disease (CVD) readmissions impose substantial clinical and economic burden. Machine learning (ML) may improve risk stratification, yet most predictive models aggregate CVD subtypes into a single outcome and underrepresent Black populations. MethodsUsing Virginia Health Information database records (2010 to 2020), we analyzed 157,791 discharge records from 123,272 unique patients (96.6% Black) to develop condition-specific 30-day readmission models for heart failure (HF; n=91,752), acute myocardial infarction (AMI; n=34,497), atrial fibrillation/flutter (AF/AFL; n=18,424), and hypertensive heart disease (HHD; n=13,118). Four algorithms (XGBoost, LightGBM, Random Forest, Elastic Net) plus a Super Learner ensemble were trained on patient-grouped 70/30 splits with and without Synthetic Minority Oversampling Technique balancing. Models incorporated validated clinical indices (LACE, Charlson, Elixhauser) and administrative social determinants of health proxies. ResultsThe overall 30-day readmission rate was 18.9%. Best area under the receiver operating characteristic curve (AUC) values by condition were HF 0.708 (95% CI, 0.701 to 0.716), AMI 0.706 (95% CI, 0.691 to 0.721), AF/AFL 0.732 (95% CI, 0.715 to 0.750), and HHD 0.758 (95% CI, 0.735 to 0.777). XGBoost was the top-performing algorithm for three of four subtypes. The LACE Index, Charlson Comorbidity Index, and insurance type were consistently the strongest predictors. Algorithm-native, aggregated, and SHAP-based importance measures converged on these key features. ConclusionsIn this largest-to-date, predominantly Black statewide cohort, condition-specific ML models achieved moderate-to-high discrimination for HF, AMI, AF/AFL, and HHD. Key clinical indices and administrative social determinants proxies emerged as dominant predictors, highlighting modifiable targets and high-risk subgroups. These findings support the development of precision, equity-informed readmission interventions and provide a scalable framework for deploying ML-driven decision support in safety-net and minority-serving healthcare systems. WHAT IS KNOWN* Machine learning models for cardiovascular readmission prediction have largely aggregated disease subtypes and underrepresented Black populations. * Most existing studies lack head-to-head algorithm comparisons within racially concentrated cohorts and omit social determinants of health proxies. WHAT THE STUDY ADDS* Condition-specific models for four cardiovascular subtypes achieved moderate-to-high discrimination (AUC 0.690 to 0.706) in the largest machine learning-based analysis of a predominantly Black statewide cohort. * Validated clinical indices (LACE, Charlson) and insurance type consistently emerged as dominant predictors, identifying modifiable targets for equity-informed intervention. * The scalable, administrative-data-only framework supports deployment of subtype-specific readmission decision support in safety-net and minority-serving health systems.

Purpose: Spatial distribution of coronary artery calcium (CAC) may provide additional prognostic value in patients undergoing SPECT and PET myocardial perfusion imaging (MPI). We aimed to automatically identify CAC in proximal segments from attenuation correction CT (CTAC) scans using artificial intelligence (AI) and to evaluate prognostic significance in two large international multicenter registries. Methods: From hybrid MPI/CT imaging (N=43,099) across 15 sites, we included 4,552 most relevant patients with 1) no prior coronary artery disease; 2) AI-derived mild CAC scores (1-99); and 3) normal perfusion (stress total perfusion deficit <5%). The independent associations between AI-identified proximal CAC and major adverse cardiovascular events (MACE) and all-cause mortality (ACM) were evaluated using multivariable Cox regression, likelihood ratio test (LRT), and continuous net reclassification index (NRI). Results: Among the patients with mild CAC and normal perfusion (mean age 65{+/-}12 years, 51% male), 1,730 (38%) had proximal CAC. Over 3.6 (inter-quartile interval 2.1, 5.2) years follow up, 599 (13%) and 444 (10%) patients had MACE or ACM, respectively. Proximal CAC was associated with an increased risk of MACE (adjusted hazard ratio [HR] 1.24, 95% CI 1.03-1.48, P=0.02) and ACM (adjusted HR 1.25, 95% CI 1.01-1.53, P=0.04) after the adjustment of CAC score and density, clinical risk factors, and perfusion deficit. Proximal CAC improved the risk stratification of MACE (LRT P=0.02; NRI 12%) and ACM (LRT P=0.04; NRI 12%). Conclusion: In patients with mild CAC and normal perfusion, AI detection of proximal CAC identified a higher-risk group for adverse outcomes, highlighting its prognostic utility.