Background:Myocardial ischemia can trigger ventricular arrhythmias with life-threatening consequences. Current monitoring is largely reactive, limiting opportunities for preventive intervention. Objective:This study aimed to determine whether high-resolution epicardial electrograms contain predictive signatures that enable forecasting the timing of premature ventricular contractions (PVCs) during acute ischemia, and to quantify subject-specific data requirements for effective personalization. Methods:We analyzed epicardial sock electrograms (247 electrodes, 1 kHz) from 21 porcine acute ischemia experiments comprising 2252 spontaneous PVCs. Signals were segmented into overlapping sequences of 3, 5, or 7 consecutive non-PVC beats with a continuous target of time-to-next PVC. A 6-layer long short-term memory network (hidden size 128) with temporal attention was trained using mean absolute error (MAE). Performance was evaluated in (A) pooled 80/10/10 cross-validation and (B) leave-1-experiment-out testing with subject-specific fine-tuning using 10% or 15% of held-out data. Results:In Paradigm A, MAE decreased with longer context (6.50 seconds for 3 beats, 5.97 seconds for 5 beats, and 4.73 seconds for 7 beats) with excellent calibration (R 2 > 0.996). In Paradigm B, increasing fine-tuning from 10% to 15% reduced mean MAE by 9.6-14.6 seconds and flattened error growth with prediction horizon, improving the fraction of predictions within 30-60 seconds windows. Conclusion:Epicardial electrograms support accurate PVC time-to-event forecasting during acute ischemia, and modest subject-specific adaptation substantially improves generalization, motivating development of real-time predictive monitoring tools.

Background:Self-supervised contrastive learning has emerged as a powerful paradigm for learning generalizable representations from unlabeled data. In the context of electrocardiogram (ECG) analysis, such pre-training can significantly enhance classification performance, especially when labeled data is scarce. Objective:We aimed to investigate and improve contrastive self-supervised learning techniques for ECGs by systematically combining recent methodological advances in augmentation design, contrastive loss formulation, and encoder architectures. Methods:We implemented a contrastive pre-training framework combining vectorcardiography (VCG)-based physiologically-inspired augmentations, interlead, intersegment, contrastive loss, and patient-aware positive sampling. In addition, we developed a dual-stream architecture, extending the TemporalNet model by processing grouped ECG leads independently. Pretraining was conducted on a large corpus of approximately 1 million unlabeled ECGs. We evaluated performance on 2 downstream classification tasks-low left ventricular ejection fraction (LVEF) and high serum potassium chloride-using various levels of labeled supervision (1%, 5%, 10%, 50%, and 100%). The pre-trained models were compared with the randomly initialized models under both frozen and finetuned conditions. Results:Contrastive pre-training consistently improved performance across all supervision levels. In low-label settings (1%-10% supervision), the pre-trained model achieved 3%-4% higher area under the receiver operator curve on the LVEF task and 5%-7% higher area under the receiver operator curve on the potassium chloride task compared with the baseline. The performance gap narrowed with increased supervision but remained favorable toward pre-trained models. Conclusion:Our findings demonstrate that contrastive pre-training can substantially enhance ECG classification, especially when labeled data is limited. By unifying and extending ideas from recent literature into a scalable framework trained on 1 million ECGs, we provide practical guidance and architectural innovations for building strong ECG foundation models applicable to a broad range of clinical prediction tasks.

Cardiovascular disease remains a major cause of morbidity and mortality, with heart failure representing the common and often irreversible end stage, for which current therapies are largely palliative rather than restorative. Cardiac microcurrent therapy delivers a low-intensity, nondepolarizing direct current across the myocardium and has been proposed to modulate fibrosis, inflammation, and remodeling in heart failure with reduced ejection fraction. In Cardiac Microcurrent I (New York Heart Association III nonischemic heart failure with reduced ejection fraction), left ventricular ejection fraction increased by 12.6% at 6 months, 6-minute walk distance improved by 191 m, and 8 of 10 patients improved to New York Heart Association I within 2 weeks. In the randomized Cardiac Microcurrent II trial, patients showed further improvements, including a 5.1% increase in left ventricular ejection fraction, a reduction in New York Heart Association class (up to 69%), and a 38 m increase in 6-minute walk distance. However, NT-proBNP (N-terminal pro-B-type natriuretic peptide) levels did not show a consistent corresponding reduction. In vitro, microcurrent altered the myofibroblast phenotype and reduced the expression of transforming growth factor-β1, collagen, α-smooth muscle actin, and fibronectin. Transcriptomics revealed downregulation of inflammatory and immune pathways, while cardiomyocytes from spontaneously hypertensive rats exhibited enhanced survival and stress resistance. This review combines early clinical observations and preclinical findings to examine potential mechanisms of action and bioelectrical effects of cardiac microcurrent therapy. Given the limited clinical evidence and largely in vitro nature of the data, the current mechanistic findings should be considered preliminary. We highlight key knowledge gaps and propose future studies to define molecular mechanisms, discover biomarkers, analyze key signaling pathways, and evaluate clinical outcomes in adequately powered, sham-controlled trials, which could further validate this therapeutic approach.

This study quantified the effect of conduction velocity (CV) variability on cardiac electrical activation patterns, a key factor for cardiac digital twins. We examined how myocardial and endocardial longitudinal, transverse, and sheet CVs influence ventricular activation across multiple pacing sites. Three porcine biventricular heart models, each including a fast-conducting endocardial layer, were used to simulate electrical activation with an eikonal approach. Uncertainty quantification with polynomial chaos expansion systematically varied six CV parameters within physiological ranges. In total, 1,868 simulations from eight ventricular pacing sites were analyzed for activation time, variability, and global sensitivities. Myocardial longitudinal CV showed the greatest influence on activation timing (global sensitivity up to 0.98). Endocardial-layer longitudinal CV was similarly important for endocardial stimuli, while transverse and sheet CVs had minimal effects. Activation-time variability reached 15 ms, increasing with distance from the pacing origin. Longitudinal CVs, particularly myocardial and endocardial-layer, dominate ventricular activation dynamics and should be prioritized when personalizing cardiac digital twins. Accounting for CV uncertainty is essential for accurate prediction and therapy optimization.

BACKGROUND:Although early studies of cardiac stereotactic body radiotherapy (cSBRT) show promise as a therapy to treat drug-refractory ventricular tachycardia, there remain knowledge gaps in its application, including the ideal doses to apply, which radiation source to use, which substrates to target, and the structural and electrophysiological effects seen after cSBRT. OBJECTIVE:Here, we examine doses up to 50 Gy to assess lesion formation by longitudinally following healthy myocardium after photon radiation using cardiac magnetic resonance (CMR) imaging and correlating those findings with histologic analysis. METHODS:Eight healthy canines underwent stereotactic photon beam cSBRT targeting healthy cardiac tissue. Serial CMR imaging assessed structural changes over time, imaging with late gadolinium enhancement, pre- and postcontrast T1, and precontrast T2 mapping. CMR images were registered to dosing plans, and left ventricular regions in 10 Gy increments from 0 to 50 Gy were identified for analysis. Histologic analysis was conducted postmortem to quantify fibrosis. RESULTS:Scar-like late gadolinium enhancement intensity was observed only in regions receiving 40 to 50 Gy within 3 months after cSBRT but was confined to small relative volumes (<4% of target volume). T1 and extracellular volume values increased in a dose- and time-dependent manner, reaching physiological levels associated with diffuse fibrosis primarily with high doses yet present in low-dose regions. Histologic examination revealed a 102.7% relative increase (+0.38% absolute increase) in diffuse fibrosis in targeted regions compared with nonirradiated controls and a 34.41% relative increase (+0.19% absolute increase) in diffuse fibrosis in nontargeted regions at chronic time points (>90 days). CONCLUSIONS:Photon cSBRT induces structural remodeling in healthy ventricular myocardium, primarily manifesting as diffuse remodeling that progressively increases with the radiation dose and the time interval after radiation without large areas of dense scar formation. These findings provide insight into the long-term effects of cSBRT, showing that healthy myocardium is unlikely to form large scar regions within 3 months and that diffuse remodeling should be further considered as a long-term effect of cSBRT.

Cardiac digital twins (CDTs) are emerging as powerful tools in personalized medicine, providing subject-specific models to simulate and understand cardiac function. A central challenge in constructing CDTs is accurately personalizing the structure and function of the His-Purkinje system (HPS), which determines ventricular activation. In this study, we leveraged a novel modified Geodesic-BP method to infer early activation sites (EASs) from epicardial activation times. The EASs can then serve as a surrogate for Purkinje-myocardial junctions, facilitating anterograde ventricular activation. We used both experimental porcine (N = 5) and synthetic (N = 5) datasets of epicardial activation times measured or assigned to locations on an electrode sock. For both datasets, we optimized for initial estimates of 5, 50, 100, and 200 EASs and assessed output variability by repeating the inference process 10 times. Assessments were based on matching the predicted activation times at both the epicardial locations and from intracardiac measurements made with multielectrode needles, and throughout the ventricular myocardium for the synthetic dataset. The algorithm could consistently recover global ventricular activation patterns from epicardial data alone. For the experimental dataset, the minimum and maximum mean absolute differences were 0.19 ms and 3.86 ms on the epicardial sock and 2.65 ms and 10.69 ms for the needles. For the synthetic dataset, the corresponding values were 0.13 ms and 2.81 ms on the sock and 2.42 ms and 14.07 ms ms throughout the ventricular myocardium. However, discrepancies between the epicardial surface and intramural myocardium, overfitting of EASs, and variability across repeated runs revealed key limitations. These findings highlight both the overall potential and current limitations of inferring EASs using the proposed optimization approach. They demonstrate the feasibility of deriving informative activation patterns from limited data, while underscoring the need to incorporate stronger physiological priors and anatomical constraints. Ultimately, our results motivate future efforts to refine simulation-based personalization frameworks, improve robustness, and enhance the physiological realism of CDTs for more accurate and reliable applications.

Ventricular arrhythmia frequently complicates myocardial ischemic events, sometimes to devastating ends. Accurate arrhythmia prediction in this setting could improve outcomes, yet traditional models struggle with the temporal complexity of the data. This study employs a Long Short-Term Memory (LSTM) network to predict the time to the next premature ventricular contraction (PVC) using high-resolution experimental data. We analyzed electrograms from 11 large animal experiments, identifying 1832 PVCs, and computed time-to-PVC. An LSTM model (247 inputs, 1024 hidden units) was trained on 10 experiments, with one held out for testing, achieving a validation MAE of 8.6 seconds and a test MAE of 135 seconds (loss 68.5). Scatter plots showed strong validation correlation and a positive test trend, suggesting the potential of this approach.

Predictive models and computational simulations of cardiac electrophysiology depend on precise anatomical representations, including the local myocardial fibre structure. However, obtaining patient-specific fibre information is challenging. In addition, the influence of physiological variability in fibre orientation on cardiac activation simulations is poorly understood. We implemented rule-based algorithms to generate fibres and robust uncertainty quantification methods to determine model output variability with respect to ventricular activation sequences. We used polynomial chaos, which reduces computational demands by using an emulator to approximate the underlying forward model. Our study examined activation sequences in response to nine stimuli and five metrics quantifying essential features of the activation sequence. The results indicated that the primary fibre orientation impacts the overall spread of activation, which could impact more complex patterns of activation; however, there is minimal impact on the location of discrete activation features, such as breakthrough sites. For free wall stimuli, the standard deviation (STD) was highest near the stimulus site, diminishing with distance. Apical stimuli showed complementary STD patterns, with epicardial pacing maximizing STD in the right basal area and endocardial pacing in the left. Ventricular junction stimuli exhibited symmetrical STD patterns, low near the stimulus but increasing sharply towards the apex, peaking on the left in the apical region. Furthermore, variability in the imbrication or helix angle did not impact the activation sequences. We conclude that in many relevant modelling contexts, the variability in myocardial fibre orientation can play an important role in the resulting activation sequences and should be accounted for. KEY POINTS: The primary fibre orientation has modest impact on activation duration and location of most discrete activation features for ectopic stimuli, but introduces variability in activation sequence. For free wall stimuli, the standard deviation (STD) was highest on the stimulated surface, indicating that deviations are largest in early activation and diminish as activation reaches remote heart regions. For apical stimuli, activation patterns were insensitive to fibre orientation variations, but the STD had complementary maxima, strongly dependent on pacing surface. Epicardial pacing produced largest STD in right basal area while endocardial pacing affected left basal area, indicating strong dependence on fibre structure. For stimuli at both anterior and posterior ventricular junctions, STD patterns were symmetrical, with low values near stimulus and sharp increases towards apex, peaking in left apical region. The helical fibre orientation showed no relevant fluctuations in activation sequence.

Identification of patient-specific scar and fibrosis is a critical step in the personalization of cardiac computational models. Late gadolinium enhanced cardiac magnetic resonance imaging (LGE-cMRI) is often used to identify patient anatomy, as well as tissue fibrosis and scar. Automated methods to identify scar from LGE-cMRI exist. Still, there is no clear consensus as to which is best in the context of patient-specific computational modeling of atrial fibrillation. There has been no substantial investigation into the effects that variability in scar may have on downstream patient-specific simulations. This study compares the distribution of scar patterns generated via automated LGE-cMRI analysis alongside human-guided scar identification. We assess the effects each identified scar pattern has on downstream computational modeling outputs by comparing the number of stable re-entrant arrhythmias induced In Silico in atrial fibrillation. We find both substantial disagreement between scar patterns identified via automated and human-guided methods, as well as sensitivity in the arrhythmia simulation outcomes across scar patterns. These results highlight the sensitivity of such computational models to these input parameters and enforce the need for robust personalization tools in the cardiac modeling field.

Background:Rotors and focal ectopies, or "drivers," are hypothesized mechanisms of persistent atrial fibrillation (AF). Machine learning algorithms have been used to identify these drivers, but the limited size of current driver data sets constrains their performance. Objective:We proposed that pretraining using unsupervised learning on a substantial data set of unlabeled electrograms could enhance classifier accuracy when applied to a smaller driver data set. Methods:We used a SimCLR-based framework to pretrain a residual neural network on 113,000 unlabeled 64-electrode measurements from a canine model of AF. The network was then fine-tuned to identify drivers from intracardiac electrograms. Various augmentations, including cropping, Gaussian blurring, and rotation, were applied during pretraining to improve the robustness of the learned representations. Results:Pretraining significantly improved driver detection accuracy compared with a non-pretrained network (80.8% vs 62.5%). The pretrained network also demonstrated greater resilience to reductions in training data set size, maintaining higher accuracy even with a 30% reduction in data. Gradient-weighted Class Activation Mapping analysis revealed that the network's attention aligned well with manually annotated driver regions, suggesting that the network learned meaningful features for driver detection. Conclusion:This study demonstrates that contrastive pretraining can enhance the accuracy of driver detection algorithms in AF. The findings support the broader application of transfer learning to other electrogram-based tasks, potentially improving outcomes in clinical electrophysiology.

Background Contractile, electrical, and structural remodeling has been associated with atrial fibrillation (AF), but the progression of functional and structural changes as AF sustains has not been previously evaluated serially. Objectives Using a rapid-paced persistent AF canine model, the authors aimed to evaluate the structural and functional changes serially as AF progresses. Methods Serial electrophysiological studies in a chronic rapid-paced canine model (n = 19) prior to AF sustaining and repeated at 1, 3, and 6 months of sustained AF were conducted to measure changes in atrial conduction speed and direction. Cardiac late gadolinium enhancement magnetic resonance imaging was performed prior to and following sustained AF to evaluate structural remodeling. Results As AF progressed, the overall area of the left atrium with fibrosis increased. Over time, conduction speeds slowed, with speeds decreasing by 0.15 m/s after 3 months and 0.26 m/s after 6 months of sustained AF. Regions that developed fibrosis experienced greater slowing compared with healthy regions (0.32 ± 0.01 m/s decrease vs 0.21 ± 0.01 m/s decrease; P < 0.001). Conduction directions became more aligned (conduction direction heterogeneity decreased from 19.7 ± 0.1° to 17.5 ± 0.1° after 6 months of sustained AF; P < 0.001). Fibrotic regions had a greater decrease in conduction direction heterogeneity (2.7 ± 0.3° vs 2.0 ± 0.2°; P = 0.008). Conclusions As AF progressed, functional changes occurred globally throughout the left atrium. Conduction speed slowed, and conduction directions became more aligned over time, with the greatest changes occurring within regions that developed fibrosis.

Background The combination of radiotherapy and immunotherapy (immunoradiotherapy) has been increasingly used for treating a wide range of cancers. However, some tumors are resistant to immunoradiotherapy. We have previously shown that MER proto-oncogene tyrosine kinase (MerTK) expressed on macrophages mediates resistance to immunoradiotherapy. We therefore sought to develop therapeutics that can mitigate the negative impact of MerTK. We designed and developed a MerTK specific antisense oligonucleotide (ASO) and characterized its effects on eliciting an anti-tumor immune response in mice. Methods 344SQR cells were injected into the right legs on day 0 and the left legs on day 4 of 8-12 weeks old female 129sv/ev mice to establish primary and secondary tumors, respectively. Radiation at a dose of 12 Gy was given to the primary tumors on days 8, 9, and 10. Mice received either anti-PD-1, anti-CTLA-4 or/and MerTK ASO starting from day 1 post tumor implantation. The composition of the tumor microenvironment and the level of MerTK on macrophages in the tumor were evaluted by flow cytometry. The expression of immune-related genes was investigated with NanoString. Lastly, the impact of MerTK ASO on the structure of the eye was histologically evaluated. Results Remarkably, the addition of MerTK ASO to XRT+anti-PD1 and XRT+anti-CTLA4 profoundly slowed the growth of both primary and secondary tumors and significantly extended survival. The ASO significantly reduced the expression of MerTK in tumor-associated macrophages (TAMs), reprograming their phenotype from M2 to M1. In addition, MerTK ASO increased the percentage of Granzyme B + CD8 + T cells in the secondary tumors when combined with XRT+anti-CTLA4. NanoString results demonstrated that the MerTK ASO favorably modulated immune-related genes for promoting antitumor immune response in secondary tumors. Importantly, histological analysis of eye tissues demonstrated that unlike small molecules, the MerTK ASO did not produce any detectable pathology in the eyes. Conclusions The MerTK ASO can significantly downregulate the expression of MerTK on TAMs, thereby promoting antitumor immune response. The combination of MerTK ASO with immunoradiotherapy can safely and significantly slow tumor growth and improve survival. Graphical Abstract

The immediate impact of catheter ablation on left atrial mechanical function and the timeline for its recovery in patients undergoing ablation for atrial fibrillation (AF) remain uncertain. The mechanical function response to catheter ablation in patients with different AF types is poorly understood. A total of 113 AF patients were included in this retrospective study. Each patient had three magnetic resonance imaging (MRI) studies in sinus rhythm: one pre-ablation, one immediate post-ablation (within 2 days after ablation), and one post-ablation follow-up MRI (≤ 3 months). We used feature tracking in the MRI cine images to determine peak longitudinal atrial strain (PLAS). We evaluated the change in strain from pre-ablation, immediately after ablation to post-ablation follow-up in a short-term study (< 50 days) and a 3-month study (3 months after ablation). The PLAS exhibited a notable reduction immediately after ablation, compared to both pre-ablation levels and those observed in follow-up studies conducted at short-term (11.1 ± 9.0 days) and 3-month (69.6 ± 39.6 days) intervals. However, there was no difference between follow-up and pre-ablation PLAS. The PLAS returned to 95

Atrial fibrillation (AF) is the most common cardiac arrhythmia and is sustained by spontaneous focal excitations and re-entry. Spontaneous electrical firing in the pulmonary vein (PV) sleeves is implicated in AF generation. The aim of this simulation study was to identify the mechanisms determining the localisation of AF triggers in the PVs and their contribution to the genesis of AF. A novel biophysical model of the canine atria was used that integrates stochastic, spontaneous subcellular Ca2+ release events (SCRE) with regional electrophysiological heterogeneity in ionic properties and a detailed three-dimensional model of atrial anatomy, microarchitecture and patchy fibrosis. Simulations highlighted the importance of the smaller inward rectifier potassium current (I-K1) in PV cells compared to the surrounding atria, which enabled SCRE more readily to result in delayed-afterdepolarisations that induced triggered activity. There was a leftward shift in the dependence of the probability of triggered activity on sarcoplasmic reticulum Ca2+ load. This feature was accentuated in 3D tissue compared to single cells (Delta half-maximal [Ca2+]SR = 58 mu M vs. 22 mu M). In 3D atria incorporating electrical heterogeneity, excitations preferentially emerged from the PV region. These triggered focal excitations resulted in transient re-entry in the left atrium. Addition of fibrotic patches promoted localised emergence of focal excitations and wavebreaks that had a more substantial impact on generating AF-like patterns than the PVs. Thus, a reduced I-K1, less negative resting membrane potential, and fibrosis-induced changes of the electrotonic load all contribute to the emergence of complex excitation patterns from spontaneous focal triggers.

The mechanisms that drive and sustain atrial fibrillation (AF) continue to be a highly researched topic. Atrial fibrosis has been linked with increased incidence of AF and conduction, but how fibrosis may lead to AF sustaining remains unknown. Our study aims to highlight heterogeneity in atrial fibrosis and how differences in fibrotic architecture may influence the sustainability of AF. In our study, we utilize a transgenic goat model with cardiac-specific over-expression of TGFβ-1 gene to examine structural differences of the fibrotic regions between animals that are inducible for AF and animals that remain AF-free. Our results indicate that there are structural differences between the fibrotic regions of AF inducible and non-inducible animals. Animals inducible for AF were found to have increased structural isotropy and increased fiber disarray within the fibrotic regions. Histology samples taken from the fibrotic regions showed fibrotic strands disrupted the tissue fibers in a more obstructive manner in the inducible animal group. These results highlight the heterogeneous differences of fibrotic regions.

Late gadolinium-enhanced (LGE) MRI has become a widely used technique to non-invasively image the left atrium prior to catheter ablation. However, LGE-MRI images are prone to variable image quality, with quality metrics that do not necessarily correlate to the image’s diagnostic quality. In this study, we aimed to define consistent clinically relevant metrics for image and diagnostic quality in 3D LGE-MRI images of the left atrium, have multiple observers assess LGE-MRI image quality to identify key features that measure quality and intra/inter-observer variabilities, and train and test a CNN to assess image quality automatically. We identified four image quality categories that impact fibrosis assessment in LGE-MRI images and trained individuals to score 50 consecutive pre-ablation atrial fibrillation LGE-MRI scans from the University of Utah hospital image database. The trained individuals then scored 146 additional scans, which were used to train a convolutional neural network (CNN) to assess diagnostic quality. There was excellent agreement among trained observers when scoring LGE-MRI scans, with inter-rater reliability scores ranging from 0.65 to 0.76 for each category. When the quality scores were converted to a binary diagnostic/non-diagnostic, the CNN achieved a sensitivity of 0.80 ± 0.06 and a specificity of 0.56 ± 0.10 . The use of a training document with reference examples helped raters achieve excellent agreement in their quality scores. The CNN gave a reasonably accurate classification of diagnostic or non-diagnostic 3D LGE-MRI images of the left atrium, despite the use of a relatively small training set.

The electrocardiogram (ECG) is the most common clinical tool to measure the electrical activity of the heart. Despite its ubiquity and utility, traditional ECG analysis methods are limited to primarily human interpretation. Machine learning tools can be employed to automate detection of diseases, and to detect patterns that are not available to traditional ECG analysis. However, contemporary machine learning tools are limited by requirements for large labeled datasets, which can be scarce for rare diseases. Self-supervised learning (SSL) can address this data scarcity. We implemented the momentum contrast (MoCo) framework, a form of SSL, using a large clinical ECG dataset. We then assessed the learning using Low Left Ventricular Ejection Fraction (LVEF) detection as the downstream task. We compared the SSL improvement of LVEF classification across different input augmentations. We observed that optimal augmentation hyperparameters varied substantially based on the training dataset size, indicating that augmentation strategies may need to be tuned based on problem and dataset size.