Atrial fibrillation (AF), the most common cardiac arrhythmia, affects one in three adults over 45 years of age. Improving its treatment requires a better understanding of bi-atrial anatomy. Existing benchmarks have focused on the left atrial (LA) cavity, overlooking the fundamental challenges posed by bi-atrial anatomy, most notably the thin atrial walls, which are critical for substrate-guided ablation planning in patients with atrial fibrillation. To address these limitations, the Multi-class Bi-Atrial Segmentation 2024 Challenge (MBAS2024) introduced the first large-scale, multi-class benchmark for simultaneous segmentation of the LA cavity, right atrial (RA) cavity, and bi-atrial walls from late gadolinium-enhanced (LGE) MRI. We systematically evaluated 13 state-of-the-art methods on the world's largest curated bi-atrial dataset, comprising 175 3D multi-center scans with expert-validated annotations, providing a comprehensive assessment of current methodological capabilities and limitations. Key findings include: segmentation of the LA and RA cavities is generally robust to image quality, whereas atrial wall delineation is highly sensitive to image degradation. Performance varies across centers, indicating limited generalization of atrial wall segmentation across different acquisition protocols. Model architecture, rather than hyperparameter tuning, is the primary driver of performance, with U-Net-based models and emerging state-space models (e.g., UMambaBot) achieving higher accuracy at modest computational cost. Segmentation accuracy also varies along the slice dimension, with central slices segmented more reliably. Finally, hybrid labeling strategies-separating LA and RA cavities while merging bi-atrial walls into a single class-consistently improve performance. The MBAS2024 challenge establishes a foundational benchmark for bi-atrial segmentation, providing validated baselines and actionable insights to guide the development of clinically relevant, efficient, and anatomically aware segmentation algorithms to improve targeted ablation in patients with AF.

The functional analysis of the left atrium (LA) is important for evaluating cardiac health and understanding diseases like atrial fibrillation. Cine MRI is ideally placed for the detailed 3D characterization of LA motion and deformation but is lacking appropriate acquisition and analysis tools. Here, we propose tools for the Analysis of Left Atrial Displacements and DeformatIons using online learning neural Networks (Aladdin) and present a technical feasibility study on how Aladdin can characterize 3D LA function globally and regionally. Aladdin includes an online segmentation and image registration network, and a strain calculation pipeline tailored to the LA. We create maps of LA Displacement Vector Field (DVF) magnitude and LA principal strain values from images of 10 healthy volunteers and 8 patients with cardiovascular disease (CVD), of which 2 had large left ventricular ejection fraction (LVEF) impairment. We additionally create an atlas of these biomarkers using the data from the healthy volunteers. Results showed that Aladdin can accurately track the LA wall across the cardiac cycle and characterize its motion and deformation. Global LA function markers assessed with Aladdin agree well with estimates from 2D Cine MRI. A more marked active contraction phase was observed in the healthy cohort, while the CVD $\text {LVEF}_{\downarrow } $ group showed overall reduced LA function. Aladdin is uniquely able to identify LA regions with abnormal deformation metrics that may indicate focal pathology. We expect Aladdin to have important clinical applications as it can non-invasively characterize atrial pathophysiology. All source code and data are available at: https://github.com/cgalaz01/aladdin_cmr_la.

In recent years, several deep learning models have been proposed to accurately quantify and diagnose cardiac pathologies. These automated tools heavily rely on the accurate segmentation of cardiac structures in MRI images. However, segmentation of the right ventricle is challenging due to its highly complex shape and ill-defined borders. Hence, there is a need for new methods to handle such structure's geometrical and textural complexities, notably in the presence of pathologies such as Dilated Right Ventricle, Tricuspid Regurgitation, Arrhythmogenesis, Tetralogy of Fallot, and Inter-atrial Communication. The last MICCAI challenge on right ventricle segmentation was held in 2012 and included only 48 cases from a single clinical center. As part of the 12th Workshop on Statistical Atlases and Computational Models of the Heart (STACOM 2021), the M&Ms-2 challenge was organized to promote the interest of the research community around right ventricle segmentation in multi-disease, multi-view, and multi-center cardiac MRI. Three hundred sixty CMR cases, including short-axis and long-axis 4-chamber views, were collected from three Spanish hospitals using nine different scanners from three different vendors, and included a diverse set of right and left ventricle pathologies. The solutions provided by the participants show that nnU-Net achieved the best results overall. However, multi-view approaches were able to capture additional information, highlighting the need to integrate multiple cardiac diseases, views, scanners, and acquisition protocols to produce reliable automatic cardiac segmentation algorithms.

Although atrial fibrillation (AF) is the most common sustained atrial arrhythmia, treatment success for this condition remains suboptimal. Information from magnetic resonance imaging (MRI) has the potential to improve treatment efficacy, but there are currently few automatic tools for the segmentation of the atria in MR images. In the study, we propose a LA-Net, a multi-task network optimised to simultaneously generate left atrial segmentation and edge masks from MRI. LA-Net includes cross attention modules (CAMs) and enhanced decoder modules (EDMs) to purposefully select the most meaningful edge information for segmentation and smoothly incorporate it into segmentation masks at multiple-scales. We evaluate the performance of LA-Net on two MR sequences: late gadolinium enhanced (LGE) atrial MRI and atrial short axis balanced steady state free precession (bSSFP) MRI. LA-Net gives Hausdorff distances of 12.43 mm and Dice scores of 0.92 on the LGE (STACOM 2018) dataset and Hausdorff distances of 17.41 mm and Dice scores of 0.90 on the bSSFP (in-house) dataset without any post-processing, surpassing previously proposed segmentation networks, including U-Net and SEGANet. Our method allows automatic extraction of information about the LA from MR images, which can play an important role in the management of AF patients.

Clinical evidence suggests a link between fibrosis in the left atrium (LA) and atrial fibrillation (AF), the most common sustained arrhythmia.Image-derived fibrosis is increasingly used for patient stratification and therapy guidance.However, locations of re-entrant drivers (RDs) sustaining AF are unknown and therapy success rates remain suboptimal.This study used image-derived LA models to explore the dynamics of RD stabilization in fibrotic regions and generate maps of RD locations.LA models with patient-specific geometry and fibrosis distribution were derived from late gadolinium enhanced magnetic resonance imaging of 6 AF patients.In each model, RDs were initiated at multiple locations, and their trajectories were tracked and overlaid on the LA fibrosis distributions to identify the most likely regions where the RDs stabilized.The simulations showed that the RD dynamics were strongly influenced by the amount and spatial distribution of fibrosis.In patients with fibrosis burden greater than 25%, RDs anchored to specific locations near large fibrotic patches.In patients with fibrosis burden below 25%, RDs either moved near small fibrotic patches or anchored to anatomical features.The patient-specific maps of RD locations showed that areas that harboured the RDs were much smaller than the entire fibrotic areas, indicating potential targets for ablation therapy.Ablating the predicted locations and connecting them to the existing pulmonary vein ablation lesions was the most effective in-silico ablation strategy. Author summaryAtrial fibrillation (AF) is the most common cardiac arrhythmia and a huge healthcare problem, but its mechanisms are incompletely understood, and clinical therapies such as catheter ablation (CA) have poor long-term outcomes.This is due to the empirical nature of the procedure and lack of mechanistic knowledge of optimal ablation sites and strategies in patients whose atria is altered by AF-induced structural remodelling.In this study,

Atrial fibrillation (AF) is a common cardiac arrhythmia that affects 1% of the population and is associated with high levels of morbidity and all-cause mortality. Catheter ablation (CA) has become one of the first line treatments for AF, but the success rates of CA and other clinical treatments remain suboptimal. The need to improve clinical outcomes warrants the optimisation of CA therapy. In this study, we develop a novel deep learning method to identify specific ablation patterns that terminate AF efficiently. To achieve this, we simulate typical AF ablation scenarios using computational models of 2D atrial tissue, and use the simulation outcomes as inputs for a deep neural network. The network is trained, validated and then applied to classify the scenarios and predict the optimal CA pattern in each scenario. For the validation dataset, the overall accuracy in identifying the best CA strategy is recorded at 79%. The study provides proof of concept that deep neural networks can learn from computational models of AF and help optimise CA therapy.

Structural changes to the wall of the left atrium are known to occur with conditions that predispose to Atrial fibrillation. Imaging studies have demonstrated that these changes may be detected non-invasively. An important indicator of this structural change is the wall's thickness. Present studies have commonly measured the wall thickness at few discrete locations. Dense measurements with computer algorithms may be possible on cardiac scans of Computed Tomography (CT) and Magnetic Resonance Imaging (MRI). The task is challenging as the atrial wall is a thin tissue and the imaging resolution is a limiting factor. It is unclear how accurate algorithms may get and how they compare in this new emerging area. We approached this problem of comparability with the Segmentation of Left Atrial Wall for Thickness (SLAWT) challenge organised in conjunction with MICCAI 2016 conference. This manuscript presents the algorithms that had participated and evaluation strategies for comparing them on the challenge image database that is now open-source. The image database consisted of cardiac CT (n=10) and MRI (n=10) of healthy and diseased subjects. A total of 6 algorithms were evaluated with different metrics, with 3 algorithms in each modality. Segmentation of the wall with algorithms was found to be feasible in both modalities. There was generally a lack of accuracy in the algorithms and inter-rater differences showed that algorithms could do better. Benchmarks were determined and algorithms were ranked to allow future algorithms to be ranked alongside the state-of-the-art techniques presented in this work. A mean atlas was also constructed from both modalities to illustrate the variation in thickness within this small cohort.

Knowledge of atrial wall thickness (AWT) has the potential to provide important information for patient stratification and the planning of interventions in atrial arrhythmias. To date, information about AWT has only been acquired in post-mortem or poor-contrast computed tomography (CT) studies, providing limited coverage and highly variable estimates of AWT. We present a novel contrast agent-free MRI sequence for imaging AWT and use it to create personalized AWT maps and a biatrial atlas. A novel black-blood phase-sensitive inversion recovery protocol was used to image ten volunteers and, as proof of concept, two atrial fibrillation patients. Both atria were manually segmented to create subject-specific AWT maps using an average of nearest neighbors approach. These were then registered non-linearly to generate an AWT atlas. AWT was 2.4 ± 0.7 and 2.7 ± 0.7 mm in the left and right atria, respectively, in good agreement with post-mortem and CT data, where available. AWT was 2.6 ± 0.7 mm in the left atrium of a patient without structural heart disease, similar to that of volunteers. In a patient with structural heart disease, the AWT was increased to 3.1 ± 1.3 mm. We successfully designed an MRI protocol to non-invasively measure AWT and create the first whole-atria AWT atlas. The atlas can be used as a reference to study alterations in thickness caused by atrial pathology. The protocol can be used to acquire personalized AWT maps in a clinical setting and assist in the treatment of atrial arrhythmias.

Anti-arrhythmic drug therapy is a frontline treatment for atrial fibrillation (AF), but its success rates are highly variable. This is due to incomplete understanding of the mechanisms of action of specific drugs on the atrial substrate at different stages of AF progression. We aimed to elucidate the role of cellular, tissue and organ level atrial heterogeneities in the generation of a re-entrant substrate during AF progression, and their modulation by the acute action of selected anti-arrhythmic drugs. To explore the complex cell-to-organ mechanisms, a detailed biophysical models of the entire 3D canine atria was developed. The model incorporated atrial geometry and fibre orientation from high-resolution micro-computed tomography, region-specific atrial cell electrophysiology and the effects of progressive AF-induced remodelling. The actions of multi-channel class III anti-arrhythmic agents vernakalant and amiodarone were introduced in the model by inhibiting appropriate ionic channel currents according to experimentally reported concentration-response relationships. AF was initiated by applied ectopic pacing in the pulmonary veins, which led to the generation of localized sustained re-entrant waves (rotors), followed by progressive wave breakdown and rotor multiplication in both atria. The simulated AF scenarios were in agreement with observations in canine models and patients. The 3D atrial simulations revealed that a re-entrant substrate was typically provided by tissue regions of high heterogeneity of action potential duration (APD). Amiodarone increased atrial APD and reduced APD heterogeneity and was more effective in terminating AF than vernakalant, which increased both APD and APD dispersion. In summary, the initiation and sustenance of rotors in AF is linked to atrial APD heterogeneity and APD reduction due to progressive remodelling. Our results suggest that anti-arrhythmic strategies that increase atrial APD without increasing its dispersion are effective in terminating AF.