BackgroundAtrial septal defects (ASD) are associated with an increased incidence of atrial arrhythmias, but their electrophysiological consequences are poorly defined. We hypothesised that conduction and repolarisation would be preferentially altered in the right atrium of ASD patients.ObjectiveTo quantify atrial conduction and repolarisation in ASD patients and determine the impact of structural remodelling on restitution properties.MethodsPatients with an ASD (n = 22) underwent bi-atrial electroanatomic mapping and quantification of effective refractory periods, longitudinal and transverse local conduction. The control group comprised 24 patients without an ASD undergoing ablation for paroxysmal AF.ResultsBipolar voltage was significantly lower in ASD patients (right atrium: 1.53 ± 0.46 mV versus 1.98 ± 0.59 mV, P = 0.017; left atrium: 1.71 ± 0.36 mV versus 2.06 ± 0.63 mV, P = 0.039). There was no significant difference in global conduction velocity in either atrium between ASD and control patients. Effective refractory periods at 600 ms were not significantly different between patient groups (right atrium: 247 ± 34.7 ms versus 224 ± 36.5 ms, P = 0.071; left atrium: 244 ± 23.9 ms versus 232 ± 40.4 ms, P = 0.29). However, both conduction and repolarisation demonstrated greater rate adaptation in ASD patients in both atria.ConclusionRight atrial remodelling, characterised by atrial dilatation and increased low voltage, is present in ASD patients. During fixed rate pacing, conduction and repolarisation properties are similar between ASD and AF patients. However, the restitution properties of both conduction and repolarisation are more pronounced in ASD than AF patients.

BACKGROUND:Voltage mapping is integral to substrate assessment for ventricular tachycardia (VT) ablation; however, the spatial extent of myocardium contributing to a recorded electrogram signal remains poorly defined. Recent preclinical data assessed the relevant field of view (FOV) of 3.5 mm and 0.167 mm2 electrodes using cardiac magnetic resonance to quantify extent of viable myocardium (VM) and found FOVs of 10 millimeters and 8 millimeters, respectively. However, this is yet to be investigated with clinical data. OBJECTIVES:This study sought to assess the FOV of 1-mm and 460-μm electrodes clinically and to evaluate the ability of cardiac magnetic resonance and cardiac computed tomography (CCT) to predict voltage amplitude. METHODS:Patients undergoing VT ablation received preprocedural late gadolinium-enhanced cardiac magnetic resonance (LGE-CMR) and CCT with extracellular volume (ECV) estimation. VM was identified using standard LGE-CMR thresholds, ECV maps were computed from CCT, and unipolar voltage was recorded during ablation procedure. VM volume and volume-weighted ECV within multisize spheres around each electrode recording site were correlated with local voltage amplitude. RESULTS:A total of 16 patients were included; 15 had imaging-derived LGE-CMR/CCT-ECV analysis and 13 underwent left ventricular endocardial voltage mapping for FOV assessment. The FOV of both electrode sizes was determined to be 13 millimeters. Of the imaging modalities assessed, LGE-derived volume of VM produced the strongest correlations with voltage (1-mm electrode: r = 0.53; P < 0.001; 460-μm electrode: r = 0.49; P < 0.001). Volume-weighted ECV demonstrated weaker correlations (r = -0.34 and -0.24; P < 0.001). CONCLUSIONS:Clinical evaluation of 1-mm and 460-μm electrodes suggests a larger FOV than preclinical investigations. This quantifies a larger scale across which myocardial viability can influence unipolar electrogram signals observed during ablation procedures and suggests that smaller electrodes can improve spatial sampling density but do not necessarily provide a distinctly more localized characterization of the electrophysiologic properties of the tissue. LGE-CMR best predicted unipolar voltage, whereas CCT-ECV performed less well. Correlations were lower than expected, indicating the need for electrophysiologic assessment alongside comprehensive imaging.

Abstract Non-valvular atrial fibrillation (AF) is associated with a five-fold increased risk of stroke, mainly due to impaired contractility of the left atrium (LA) leading to blood stasis and subsequent thrombus formation within the left atrial appendage (LAA). Current AF stroke risk stratification schemes, such as the CHA₂DS₂-VASc/ CHA₂DS₂-VA score, use comorbidities and do not capture mechanistic factors like blood flow dynamics and hypercoagulability. To address this, we developed a multiphase computational fluid dynamics (CFD) model of the LA, incorporating patient-specific geometries; modelling of the coagulation cascade; and non-Newtonian blood behaviour within the LAA. Using 84 simulation cases generated via Latin Hypercube Sampling of physiological blood parameters and 21 patient-derived LA anatomies, we trained surrogate machine learning models, including Ridge regression, XGBoost, Gaussian Process Emulators (GPEs), and deep learning networks, to predict CFD outputs such as blood viscosity in and fibrin concentrations in the LAA. Deep learning achieved R² values up to 0.90, with the accuracy increasing when both physiological parameters and the raw CT image were included. Other models showed uneven performance with R 2 values below 0.7, highlighting the role of nonlinearities between parameters. The study presents a novel CFD model that captures the transition from blood stasis to clot formation, representing the full thrombotic continuum underlying stroke risk in AF, and a deep learning approach to enable efficient prediction of mechanistic outputs of clinical value for stroke risk stratification in AF patients. Author Summary Atrial fibrillation is a common heart rhythm disorder that greatly increases the risk of stroke. In many patients, blood can pool inside a small pouch of the heart called the left atrial appendage, where clots may form and later travel to the brain. Current clinical tools used to estimate stroke risk mainly rely on a patient’s medical history and do not directly assess the mechanistic processes that lead to clot formation. In this study, we developed a computer model that simulates how blood flows and clots inside the heart using patient-specific heart anatomies derived from medical imaging. Our model combines blood flow, blood biochemistry, and the changing physical properties of blood during clot formation. We then used machine learning methods to predict these complex simulation results more efficiently. Deep learning models performed best, particularly when both clinical parameters and heart imaging data were included. Our work provides a new way to study the full process linking abnormal blood flow to clot formation in atrial fibrillation. In the future, this approach could support more personalised and mechanistic assessment of stroke risk and help guide treatment decisions.

Hypertrophic cardiomyopathy is the most common inherited cardiac condition, associated with increased risks of heart failure, arrhythmias, and sudden cardiac death. Once considered a disease of the ventricular myocardium, growing evidence points towards atrial involvement, with structural and electrical atrial remodelling driven by a complex interplay of genetic, electrophysiological and hemodynamic factors. Despite advances in the understanding of this atrial substrate, management of atrial arrhythmias in hypertrophic cardiomyopathy remains challenging. Pharmacological therapies are limited by safety concerns, and catheter ablation shows lower efficacy in patients with hypertrophic cardiomyopathy compared to patients with structurally normal hearts. This review synthesizes current knowledge of disease related changes underlying atrial arrhythmias in hypertrophic cardiomyopathy and outlines diagnostic and therapeutic strategies. Particular emphasis is given to interventional treatment approaches, including emerging technologies and their evolving role in patients with hypertrophic cardiomyopathy.

Atrial fibrillation (AF) is the most common cardiac arrhythmia and increases stroke risk and reduces quality of life. Left atrial catheter ablation (LACA) restores sinus rhythm by targeting abnormal electrical sources. Pulmonary vein isolation (PVI) is the most common approach, while left atrial appendage electrical isolation (LAAEI) is used selectively but may raise thrombus risk. Computational fluid dynamics (CFD) has been used to model thrombogenesis during LACA, but the relationship between blood temperature and coagulation is yet to be modelled. We performed twenty-nine simulations coupling CFD with heat transfer during PVI and related thrombogenesis in the left atrium. These included PVI at 80 °C and 60 °C, LAAEI at 80 °C, and ablation sites at increasing distances from the LAA. Experimental data linking temperature and coagulation were used to calibrate a thermal exposure model, alongside transport and reaction equations for coagulation proteins, including fibrin. At 80 °C, PVI produced variable fibrin formation across the PVs, with no significant differences between PVs or correlation with PV velocity. At 60 °C, fibrin remained below the thrombus threshold, and fibrin at 80 °C was significantly greater (p = 1.91×10⁻⁶). Including LAAEI produced a significant overall difference in fibrin among sites, but no pairwise differences remained after correction. Fibrin decreased with increasing distance from the LAA orifice, showing a significant negative correlation after three and twelve cycles. These results highlight risks of temperature-dependent thrombogenesis and suggest reduced thrombus risk with more distant ablation from the LAA.

Background Coronary CT angiography provides prognostic information in addition to coronary findings. Purpose To evaluate associations between machine learning-derived multiorgan body composition and 10-year outcomes in the SCOT-HEART (Scottish Computed Tomography of the Heart) trial. Materials and Methods Wide field-of-view images of 1722 patients (recruited between November 2010 and September 2014) were retrospectively processed using the TotalSegmentator model. The volume and mean attenuation of segmented organs were calculated. Multivariable Cox proportional hazards models were constructed for all-cause mortality and myocardial infarction (MI), adjusted for age, sex, and scan length. Odds ratios or hazard ratios (HRs) and 95% CIs were calculated per 10-unit increase in attenuation or volume. Results Mortality and MI occurred in 133 (7.72%) and 106 (6.16%) of the 1722 patients, respectively (age, 57.5 years ± 9.5 [SD]; 55.7% male). Coronary artery disease was associated with greater lung attenuation (odds ratio, 1.04 [95% CI: 1.03, 1.06]; P < .001), lower liver attenuation (odds ratio, 0.87 [95% CI: 0.8, 0.95]; P = .034), and greater torso fat volume (odds ratio, 1.01 [95% CI: 1.01, 1.02]; P < .001) after multivariable adjustment. Increased skeletal muscle attenuation was associated with lower all-cause mortality (HR, 0.61 [95% CI: 0.47, 0.79]; P < .001) after multivariable adjustment. MI was associated with increased myocardial volume (HR, 1.09 [95% CI: 1.01, 1.16]; P = .018) and decreased rib (HR, 0.98 [95% CI: 0.96, 1.0]; P = .043) and skeletal muscle (HR, 0.69 [95% CI: 0.54, 0.87]; P = .002) attenuation after multivariable adjustment. However, when further adjusted for coronary calcium score, only skeletal muscle attenuation was associated with MI (HR, 0.72 [95% CI: 0.57, 0.91]; P = .007). Patients with skeletal muscle attenuation below the median had a higher risk of mortality (HR, 1.85 [95% CI: 1.30, 2.64]; P < .001) or experience MI (HR, 1.58 [95% CI: 1.07, 2.33]; P = .022). Conclusion Multiorgan body composition analysis using coronary CT angiography provided additional prognostic information, among which skeletal muscle attenuation was particularly important. ClinicalTrials.gov identifier: NCT01149590 © RSNA, 2026 Supplemental material is available for this article.

In atrial fibrillation (AF), atrial biomechanics are altered, reducing atrial movement. It remains unclear whether these changes are due to altered anatomy, myocardial stiffness, or constraints from surrounding structures. Understanding the causes of changed atrial deformation in AF could enhance tissue characterization and inform AF diagnosis, stratification, and treatment. We created patient-specific anatomical models of the left atrium (LA) from CT images. Passive LA biomechanics were simulated using finite deformation continuum mechanics equations. LA stiffness was represented by the Guccione material law, where α scaled the anisotropic stiffness parameters. Regional passive stiffness parameters were calibrated to peak regional deformation during the reservoir phase and validated against deformation transients derived from retrospective gated CT images during the reservoir and conduit phase. Physiological LA deformation varies regionally, with the roof deforming significantly less than other regions during the reservoir phase. The fitted model matched peak patient deformations globally and regionally with an average error of 0.90 ± 0.39 mm over our cohort. We compared deformation transients through the reservoir and conduit phases and found that the simulated deformation transients were within an average of ± 0.38 mm per unit time of the CT-derived deformation transients. Regional stiffness varied across the atria with average α values of 1.8, 1.6, 2.2, 1.6 and 2.1 across the cohort in the anterior, posterior, septum, lateral and roof regions respectively. Using mixed effect models, we found no correlation between regional patient LA deformation and regional estimates of wall thickness or regional volumes of epicardial adipose tissue. We found a significant correlation between regionally calibrated stiffness and CT-derived LA biomechanics ( p = 0.023). We have shown that regional heterogeneity in stiffness contributes to regional LA biomechanics, while anatomical features appeared less important. These findings provide insight into the underlying causes of altered LA biomechanics in AF.

BACKGROUND:Stress perfusion cardiovascular magnetic resonance (CMR) in the presence of atrial fibrillation (AF) has long been challenging due to electrocardiogram (ECG) mis-triggering. However, non-invasive ischemia imaging is important due to an increased risk of myocardial infarction in patients with AF, which has been attributed to underlying microvascular dysfunction. Myocardial blood flow (MBF) in patients with AF is poorly understood, and few studies have attempted to quantify this through non-invasive imaging. METHODS:Patients were recruited for stress perfusion CMR using a research sequence at 3-Tesla. Image acquisition occurred during both vasodilator-induced hyperemia and at rest. Stress and rest MBF maps were automatically generated. Analysis of perfusion maps included assessment of myocardial perfusion reserve (MPR) and endocardial-to-epicardial MBF ratios. RESULTS:Around 442 patients were analyzed; 63 of whom had a history of AF and were in AF during the scan. Both MBF during hyperemia (stress MBF) and MPR were reduced in patients with AF compared to those in sinus rhythm (median stress MBF 1.85 [1.52-2.24] vs. 2.35 [1.98-2.77] mL/min/g, p<0.001; median MPR 1.95 [1.62-2.19] vs. 2.37 [2.05-2.80], p<0.001). No significant difference was seen between the two groups at rest (p=0.451). When considering co-factors affecting MBF, multivariate linear regression analysis identified the presence of AF as a significant independent contributor to stress MBF and MPR values. Both endocardial and epicardial stress MBF and MPR were reduced in AF compared with sinus rhythm (both p<0.001) and endocardial/epicardial ratios were similar between the groups. CONCLUSION:Automated quantitative MBF assessment can be performed in patients with AF. At hyperemia, MBF is reduced in AF compared to sinus rhythm.

Objectives An image-based ECG dataset incorporating visual imperfections common to paper-based ECGs, which are typically scanned or photographed into electronic health records, could facilitate clinically useful artificial intelligence (AI)-ECG algorithm development. This study aimed to create a high-fidelity, synthetic image-based ECG dataset.Methods ECG images were recreated from the PTB-XL database, a signal-based dataset and image manipulation techniques were applied to mimic imperfections associated with ECGs in real-world settings. Clinical Turing tests were conducted to evaluate the fidelity of the synthetic images, and the performance of current AI-ECG algorithms was assessed using synthetic images containing visual imperfections.Results GenECG, an image-based dataset containing 21 799 ECGs with visual imperfections encountered in routine clinical care paired with imperfection-free images, was created. Turing tests confirmed the realism of the images: expert observer accuracy of discrimination between real-world and synthetic ECGs fell from 63.9% (95% CI 58.0% to 69.8%) to 53.3% (95% CI 48.6% to 58.1%) over three rounds of testing, indicating that observers could not distinguish between synthetic and real ECGs. The performance of pre-existing algorithms on synthetic (area under the curve (AUC) 0.592, 95% CI 0.421 to 0.763) and real-world (AUC 0.647, 95% CI 0.520 to 0.774) ECG images containing imperfections was limited. Algorithm fine-tuning with GenECG data improved real-world ECG classification accuracy (AUC 0.821, 95% CI 0.730 to 0.913) demonstrating its potential to augment image-based algorithm development.Discussion/conclusion GenECG is the first synthetic image-based ECG dataset to pass a clinical Turing test. The dataset will enable image-based AI-ECG algorithm development, ensuring utility in low resource areas, prehospital settings and hospital environments where signal data are unavailable.

Atrial fibrillation (AF), impacting nearly 50 million individuals globally, is a major contributor to ischaemic strokes, predominantly originating from the left atrial appendage (LAA). Current clinical scores like CHA₂DS₂-VASc, while useful, provide limited insight into the pro-thrombotic mechanisms of Virchow's triad—blood stasis, endothelial damage, and hypercoagulability. This study leverages biophysical computational modelling to deepen our understanding of thrombogenesis in AF patients. Utilising high temporal resolution Cine magnetic resonance imaging (MRI), a 3D patient-specific modelling pipeline for simulating patient-specific flow in the left atrium was developed. This computational fluid dynamics (CFD) approach was coupled with reaction-diffusion-convection equations for key clotting proteins, leading to an innovative risk stratification score that combines clinical and modelling data. This approach categorises thrombogenic risk into low (A), moderate (B), and high (C) levels. Applied to a cohort of nine patients, pre- and post-catheter ablation therapy, this approach generates novel risk scores of thrombus formation, which are based of mechanistic characterisation of all aspects of the Virchow's triad. Currently, thrombogenesis mechanisms are not factored in widespread clinical risks scores based on demographic characteristics and co-morbidities. Notably, some patients with a CHA₂DS₂-VASc score of 0 (lowest clinical risk) exhibited much higher risks once the individual pathophysiology was accounted for. This discrepancy highlights the limitations of the CHA₂DS₂-VASc score in providing detailed mechanistic insights into patient-specific thrombogenic risk. This work introduces a comprehensive method for assessing thrombus formation risks in AF patients, emphasising the value of integrating biophysical modelling with clinical scores to enhance personalised stroke prevention strategies.

The left atrium (LA) is particularly susceptible to blood stasis in conditions like atrial fibrillation (AF), which can lead to thrombus formation, especially in the left atrial appendage. Spontaneous echo contrast (SEC) in the LA, detectable via transesophageal echocardiography (TEE), occurs when blood flow slows, and has been strongly associated with thrombus formation and increased stroke risk, making it an important prognostic indicator. The underlying mechanism of LA SEC is thought to involve echogenic red blood cell aggregates formed due to low shear rates, but the roles of platelets and the coagulation cascade remain unclear. Given that LA SEC is considered a precursor to thrombus formation, enhancing our understanding of its pathophysiology may offer insights into thrombogenesis inside the LA, which to date remains poorly understood. The development of noninvasive diagnostic tools for LA SEC is critical, as TEE, whereas the gold standard is invasive and not universally accessible. Promising alternatives, such as harmonic transthoracic echocardiography and biphasic computed tomography imaging, have shown potential in diagnosing LA SEC and assessing stroke risk in AF patients. Additionally, emerging technologies like computational modelling are offering new avenues for understanding the mechanisms of LA SEC, with blood flow simulations providing valuable insights into its formation. These advancements could improve diagnostic capabilities and stroke risk stratification in AF patients, highlighting the need for further research to fully elucidate the clinical implications of LA SEC.

Atrial Fibrillation (AF) significantly increases the risk of ischemic stroke due to blood stasis and hypercoagulability in the left atrium (LA). Effective stroke risk stratification is crucial for identifying AF patients who require anticoagulation therapy. Spontaneous echo contrast (SEC), a phenomenon arising from blood stasis and fibrinogen-mediated red blood cell aggregation, serves as a strong predictor of stroke and can be observed via transesophageal echocardiograms (TEEs). This study employs Gaussian Process Emulators (GPEs) trained on non-Newtonian Computational Fluid Dynamics (CFD) simulations to predict the occurrence and intensity of LA SEC. Using haematocrit and fibrinogen concentration as inputs, the GPEs compute the average blood viscosity in the left atrial appendage (LAA). We also quantified geometric and functional parameters of the LAA to assess their relative impact on SEC intensity and compared these findings with those in patients without SEC. Our results indicate that both LAA motility and GPE-predicted LAA viscosity can distinguish between SEC and non-SEC groups (p > 0.001). In the SEC cohort, significant correlations were observed between grayscale intensity and LAA motility (r = 0.8), LAA orifice diameter (r = 0.71), and predicted LAA viscosity (r = 0.57). This study demonstrates the potential utility of GPEs for predicting LA SEC, thereby enhancing stroke risk stratification in AF patients. Additionally, we identify key geometric and functional features of the LAA that influence SEC formation.

INTRODUCTION:Measuring conduction velocity, as a direct consequence of fibrosis, may provide a better method to localise fibrotic regions. This study aims to assess established cardiac conduction velocity calculation methods (Triangulation, Polynomial Surface Fitting, and Radial Basis Function) in identifying areas of conduction slowing caused by fibrosis, considering realistic measurement errors. METHOD:Using a human left atrium computational model, atrial activation was simulated. Each conduction velocity calculation method's performance was evaluated under uncertainties in mapping point density, local activation time assignment and electrode locations by comparing calculated conduction velocity to ground truth conduction velocity derived from high-resolution simulated atrial activation. RESULTS:All methods agreed well with ground truth conduction velocity maps in noise-free, high-density sampling conditions. However, Triangulation and Polynomial Surface Fitting methods showed susceptibility to noise, exhibiting significant errors under moderate to high noise levels. Radial Basis Function method demonstrated greater robustness to noise and reduced sampling density. Fibrotic region identification accuracy was high under ideal conditions for all methods but declined with increasing noise, with the Radial Basis Function method maintaining superior performance. CONCLUSION:While all methods accurately estimate conduction velocity under ideal conditions, the Radial Basis Function method shows robustness to a realistic clinical noise, hence making it the most suitable to identify fibrotic regions.

Objective: This study aimed to develop and evaluate the SeptalPro system, a novel robotic platform designed to enhance the precision, safety, and efficiency of transseptal puncture (TSP) procedures by integrating real-time remote control and force sensing. Methods: The SeptalPro system was assessed using an anthropomorphic phantom in a simulated clinical environment. Four experienced cardiologists and one cardiology trainee participated in the study, performing manual and robot-assisted TSP procedures. The system's performance was evaluated using trajectory and motion smoothness metrics, as well as user feedback on the system's usability in a clinical environment. Results: The robot-assisted approach demonstrated superior spatial control compared to the manual approach, with significantly shorter path lengths and reduced spatial dispersion for experienced operators. The SeptalPro system also achieved smoother and more efficient tool motions, as indicated by reduced mean motion jerk. User feedback indicated high satisfaction with the system's usability and potential clinical benefits. Conclusion: SeptalPro represents a significant advancement in TSP, offering a promising solution to the challenges associated with the conventional manual approach. The robotic system demonstrated superior performance in terms of spatial control and motion smoothness compared to the manual approach. Significance: This work establishes the first comprehensive framework for evaluating robotic TSP systems through integration of high-precision force sensing, clinically-validated performance metrics, and structured assessment protocols, providing essential foundations towards clinical translation of robotic assistance in TSP procedures.

BACKGROUND:Peri-atrial adipose tissue is associated with atrial fibrillation (AF). Increased peri-atrial adipose volume and attenuation, detected by cardiac computed tomography angiography (CTA), have been observed in patients with AF. However, the electrophysiological correlates of both peri-atrial adipose tissue volume and attenuation are unknown. OBJECTIVES:This study sought to investigate the spatial relationship between peri-atrial adipose tissue, peri-atrial adipose tissue attenuation, and atrial electrophysiological remodeling. METHODS:Cardiac CTA was performed in 37 control subjects and 44 patients with AF. Left atrial bipolar voltage and conduction velocity were co-registered with cardiac CTA-derived peri-atrial adipose tissue segmentations. Mean adipose tissue volume and attenuation were compared with local voltage and conduction velocity measurements. RESULTS:Peri-atrial adipose tissue volume was greater in patients with AF (20.9 cm3 vs 14.2 cm3; adjusted odds ratio: 1.11; 95% CI: 1.01-1.24), independent of left atrial volume indexed to body mass index, left atrial mass, age, sex, sleep apnea, and coronary heart disease. In patients with AF, areas with the highest burden of peri-atrial adipose tissue had lower voltage (1.75 ± 1.72 mV vs 2.11 ± 2.02 mV; P < 0.001) and conduction velocity (0.627 ± 0.55 ms-1 vs 0.683 ± 0.48 ms-1; P < 0.001), compared with areas with the lowest burden of peri-atrial adipose tissue. Mean peri-atrial adipose tissue attenuation was similar in both groups. In patients with AF, low peri-atrial adipose tissue attenuation was weakly correlated with reduced bipolar voltage (1.69 ± 1.68 mV vs 2.16 ± 2.07 mV; P < 0.001) and conduction velocity (0.615 ± 0.47 ms-1 vs 0.684 ± 0.43 ms-1; P < 0.001). CONCLUSIONS:Peri-atrial adipose tissue volume was greater in patients with AF. Increased peri-atrial adipose tissue burden and reduced attenuation were spatially but weakly correlated with adverse electrophysiological remodeling in patients with AF.

Cardiac magnetic resonance (CMR) imaging and computed tomography (CT) are two common non-invasive imaging methods for assessing patients with cardiovascular disease. CMR typically acquires multiple sparse 2D slices, with unavoidable respiratory motion artefacts between slices, whereas CT acquires isotropic dense data but uses ionising radiation. In this study, we explore the combination of Slice Shifting Algorithm (SSA), Spatial Transformer Network (STN), and Label Transformer Network (LTN) to: 1) correct respiratory motion between segmented slices, and 2) transform sparse segmentation data into dense segmentation. All combinations were validated using synthetic motion-corrupted CMR slice segmentation generated from CT in 1699 cases, where the dense CT serves as the ground truth. In 199 testing cases, SSA-LTN achieved the best results for Dice score and Hausdorff distance ( 94.0% and 4.7 mm respectively, average over 5 labels) but gave topological errors in 8 cases. STN was effective as a plug-in tool for correcting all topological errors with minimal impact on overall performance ( 93.5% and 5.0 mm respectively). SSA also proves to be a valuable plug-in tool, enhancing performance over both STN-based and LTN-based models. The code for these different combinations is available at https://github.com/XESchong/STACOM2024 .

Reliable outcome prediction following atrial fibrillation (AF) catheter ablation is important to inform shared decision-making. The extent of atrial electromechanical remodelling can be determined through magnetic resonance imaging and electroanatomic mapping data analysis. Combining these data with clinical data could improve the accuracy of outcome prediction models. To investigate how left atrial electromechanical remodelling data can be utilised to predict outcomes for first-time and repeat AF ablation. A retrospective analysis of 123 patients undergoing first-time ablation was conducted. Clinical, imaging and electroanatomic mapping variables associated with arrhythmia recurrence were identified using univariable logistic regression and combined into a multivariable model. Predictive ability for treatment response was examined using receiver-operator characteristic curve, time-to-event analyses and compared to pre-existing clinical risk scores. A multivariable model comprising age, weight, hypertension, left atrial ejection fraction and mean left atrial voltage attained a c-statistic of 0.733 (95

Abstract Background Transseptal puncture (TSP) is a critical prerequisite for left-sided cardiac interventions, such as atrial fibrillation (AF) ablation and left atrial appendage closure. Despite its routine nature, TSP can be technically demanding and carries a risk of complications. This study presents a novel, patient-specific, anthropomorphic phantom for TSP simulation training that can be used with X-ray fluoroscopy and ultrasound imaging. Methods The TSP phantom was developed using additive manufacturing techniques and features a replaceable fossa ovalis (FO) component to allow for multiple punctures without replacing the entire model. Four cardiologists and one cardiology trainee performed TSP on the simulator, and their performance was assessed using four metrics: global isotropy index, distance from the centroid, time taken to perform TSP, and a set of 5-point Likert scale questions to evaluate the clinicians’ perception of the phantom’s realism and utility. Results The results demonstrate the simulator’s potential as a training tool for interventional cardiology, providing a realistic and controllable environment for clinicians to refine their TSP skills. Experienced cardiologists tended to cluster their puncture points closer to regions of the FO associated with higher global isotropy index scores, indicating a relationship between experience and optimal puncture localization. The questionnaire analysis revealed that participants generally agreed on the phantom’s realistic anatomical representation and ability to accurately visualize the TSP site under fluoroscopic guidance. Conclusions The TSP simulator can be incorporated into training programs, offering trainees the opportunity to improve tool handling, spatial coordination, and manual dexterity prior to performing the procedure on patients. Further studies with larger sample sizes and longitudinal assessments are needed to establish the simulator’s impact on TSP performance and patient outcomes.