Deep Learning Framework for Real-Time Estimation of in-silico Thrombotic Risk Indices in the Left Atrial Appendage

Unveiling the potential of artificial intelligence in revolutionizing disease diagnosis and prediction: a comprehensive review of machine learning and deep learning approaches

The rapid advancement of Machine Learning (ML) and Deep Learning (DL) technologies has revolutionized healthcare, particularly in the domains of disease prediction and diagnosis. This study provides a comprehensive review of ML and DL applications across sixteen diverse diseases, synthesizing findings from research conducted between 2015 and 2024. We explore these technologies' methodologies, effectiveness, and clinical outcomes, highlighting their transformative potential in healthcare settings. Although ML and DL demonstrate remarkable accuracy and efficiency in disease prediction and diagnosis, challenges including quality of data, interpretability of models, and their integration into clinical workflows remain significant barriers. By evaluating advanced approaches and their outcomes, this review not only underscores the current capabilities of ML and DL but also identifies key areas for future research. Ultimately, this work aims to serve as a roadmap for advancing healthcare practices, enhancing clinical decision making, and strengthening patient outcomes through the effective and responsible implementation of AI-driven technologies.

Super-resolution left ventricular flow and pressure mapping by Navier-Stokes-informed neural networks

Intraventricular vector flow mapping (VFM) is an increasingly adopted echocardiographic technique that derives time-resolved two-dimensional flow maps in the left ventricle (LV) from color-Doppler sequences. Current VFM models rely on kinematic constraints arising from planar flow incompressibility. However, these models are not informed by crucial information about flow physics; most notably the forces within the fluid and the resulting accelerations. This limitation has rendered VFM unable to combine information from different time frames in an acquisition sequence or derive fluctuating pressure maps. In this study, we leveraged recent advances in artificial intelligence (AI) to develop AI-VFM, a vector flow mapping modality that uses physics-informed neural networks (PINNs) encoding mass conservation and momentum balance inside the LV, and no-slip boundary conditions at the LV endocardium. AI-VFM recovers the flow and pressure fields in the LV from standard echocardiographic scans. It performs phase unwrapping and recovers flow data in areas without input color-Doppler data. AI-VFM also recovers complete flow maps at time points without color-Doppler input data, producing super-resolution flow maps. We validate AI-VFM using physiological simulated LV data and show that informing the PINNs with momentum balance is essential for achieving temporal super-resolution and significantly increases the accuracy of AI-VFM compared to informing the PINNs only with mass conservation. AI-VFM is solely informed by each patient's flow physics; it does not utilize explicit smoothness constraints or incorporate data from other patients or flow models. AI-VFM takes 15 min to run in off-the-shelf graphics processing units and its underlying PINN framework could be extended to map other flow-associated metrics such as blood residence time or the concentration of coagulation species.

Developing a nearly automated open-source pipeline for conducting computational fluid dynamics simulations in anterior brain vasculature: a feasibility study

Intracranial aneurysms (IA) pose significant health risks and are often challenging to manage. Computational fluid dynamics (CFD) simulation has emerged as a powerful tool for understanding lesion-specific hemodynamics in and around IAs, aiding in the clinical management of patients with an IA. However, the current workflow of CFD simulations is time-consuming, complex, and labor-intensive and, thus, does not fit the clinical environment. To address these challenges, we have developed a semi-automated pipeline integrating multiple open-source software packages to streamline the CFD simulation process. Specifically, the study utilized medical angiography data from 18 patients. An in-house open-source DL image segmentation model (ARU-Net) was employed to generate 3D computer models of the anterior circulation. The segmented intracranial vasculature models, including IAs, were further refined using the Vascular Modeling Toolkit (VMTK), an open-source Python package. This step involved smoothing the surface of the models and extending the inlet and outlet regions to ensure a realistic representation of the vascular geometry. The refined vascular models were then converted into computational meshes using an open-source mesh generator known as TetGen. This process was nearly automated and required minimal user interaction(s). Blood flow simulations of the cerebral vascular models were performed using established SimVascular solvers (an open-source finite element platform for vascular applications) through an application programming interface (API). The CFD simulation process was also conducted using the manual workflow for comparative purposes. The initial assessment compared the geometries derived from manual and DL-based segmentation. The DL-based segmentation demonstrated reliable performance, closely aligning with manually segmented results, evidenced by excellent Pearson correlation coefficient (PCC) values and low relative difference (RD) values ranging from 3% to 10% between the computed geometrical variables derived from both methods. The statistical analysis of the computed hemodynamic variables, including velocity informatics and WSS-related variables, indicated good to excellent reliability for most parameters (e.g., ICC of 0.85-0.95). Given the data investigated, the proposed automated workflow streamlines the process of conducting CFD simulations. It generates results consistent with the current standard manual CFD protocol while minimizing dependence on user input.

A digital twin approach for stroke risk assessment in Atrial Fibrillation Patients

Atrial fibrillation (AF) is associated with a fivefold increased risk of cerebrovascular events, contributing to 15-18 % of all strokes. Stroke prevention in clinical practice is typically guided by the CHA2DS2-VASc score, which depends on general clinical risk factors but falls short in predicting risk at an individual patient level. In this study, we introduce a digital twin model of the left atrium (LA) combined with computational fluid dynamics (CFD) simulations to enhance personalized stroke risk assessment. Simulations were performed on patient-specific dynamic LA models in sinus rhythm (SR) across three patient groups: 10 controls (CTRL), 10 with paroxysmal AF (PAR-AF), and 10 with persistent AF (PER-AF). Blood flow velocity and areas susceptible to thrombogenesis, based on several factors including endothelial damage, were identified in the left atrial appendage (LAA). In general, control subjects exhibited higher average blood velocity in both the LAA and its ostium (0.11 ± 0.03 m/s and 0.28 ± 0.05 m/s, respectively) compared to those with AF. In the AF groups, the velocities were lower (LAA: PAR-AF 0.05 ± 0.02 m/s, PER-AF 0.04 ± 0.02 m/s; LAA ostium: PAR-AF 0.14 ± 0.03 m/s, PER-AF 0.11 ± 0.04 m/s). CFD analysis revealed that endothelial cell activation potential (ECAP) was significantly higher in AF patients (PAR-AF: 3.96 ± 3.28 Pa⁻1, PER-AF: 4.77 ± 2.08 Pa⁻1) compared to controls (0.93 ± 0.63 Pa⁻1). These findings suggest that AF patients experience slower and more oscillatory blood flow in the LAA, increasing their risk of thrombosis. Additionally, blood tends to stagnate within the LAA, further raising the likelihood of clot formation. This proposed method could be used to enhance stroke risk stratification in AF patients by incorporating an index that integrates blood velocity-derived parameters.

A comprehensive stroke risk assessment by combining atrial computational fluid dynamics simulations and functional patient data

Stroke, a major global health concern often rooted in cardiac dynamics, demands precise risk evaluation for targeted intervention. Current risk models, like theCHA 2 DS 2 -VASc score, often lack the granularity required for personalized predictions. In this study, we present a nuanced and thorough stroke risk assessment by integrating functional insights from cardiac magnetic resonance (CMR) with patient-specific computational fluid dynamics (CFD) simulations. Our cohort, evenly split between control and stroke groups, comprises eight patients. Utilizing CINE CMR, we compute kinematic features, revealing smaller left atrial volumes for stroke patients. The incorporation of patient-specific atrial displacement into our hemodynamic simulations unveils the influence of atrial compliance on the flow fields, emphasizing the importance of LA motion in CFD simulations and challenging the conventional rigid wall assumption in hemodynamics models. Standardizing hemodynamic features with functional metrics enhances the differentiation between stroke and control cases. While standalone assessments provide limited clarity, the synergistic fusion of CMR-derived functional data and patient-informed CFD simulations offers a personalized and mechanistic understanding, distinctly segregating stroke from control cases. Specifically, our investigation reveals a crucial clinical insight: normalizing hemodynamic features based on ejection fraction fails to differentiate between stroke and control patients. Differently, when normalized with stroke volume, a clear and clinically significant distinction emerges and this holds true for both the left atrium and its appendage, providing valuable implications for precise stroke risk assessment in clinical settings. This work introduces a novel framework for seamlessly integrating hemodynamic and functional metrics, laying the groundwork for improved predictive models, and highlighting the significance of motion-informed, personalized risk assessments.

Super-resolution Left Ventricular Flow and Pressure Mapping by Navier-Stokes-Informed Neural Networks

Intraventricular vector flow mapping (VFM) is a growingly adopted echocardiographic modality that derives time-resolved two-dimensional flow maps in the left ventricle (LV) from color-Doppler sequences. Current VFM models rely on kinematic constraints arising from planar flow incompressibility. However, these models are not informed by crucial information about flow physics; most notably the pressure and shear forces within the fluid and the resulting accelerations. This limitation has rendered VFM unable to combine information from different time frames in an acquisition sequence or derive fluctuating pressure maps. In this study, we leveraged recent advances in artificial intelligence (AI) to develop AI-VFM, a vector flow mapping modality that uses physics-informed neural networks (PINNs) encoding mass conservation and momentum balance inside the LV, and no-slip boundary conditions at the LV endocardium. AI-VFM recovers the flow and pressure fields in the LV from standard echocardiographic scans. It performs phase unwrapping and recovers flow data in areas without input color-Doppler data. AI-VFM also recovers complete flow maps at time points without color-Doppler input data, producing super-resolution flow maps. We show that informing the PINNs with momentum balance is essential to achieving temporal super-resolution and significantly increases the accuracy of AI-VFM compared to informing the PINNs only with mass conservation. AI-VFM is solely informed by each patient's flow physics; it does not utilize explicit smoothness constraints or incorporate data from other patients or flow models. AI-VFM takes 15 minutes to run in off-the-shelf graphics processing units and its underlying PINN framework could be extended to map other flow-associated metrics like blood residence time or the concentration of coagulation species.

The impact of contrast retention on thrombus formation risks in patients with atrial fibrillation: A numerical study

Background

Contrast retention (CR) is an important predictor of left atrial appendage thrombus (LAAT) and stroke in patients with non-valvular atrial fibrillation (AF). We sought to explore the underlying mechanisms of CR using computational fluid dynamic (CFD) simulations.

Methods

A total of 12 patients with AF who underwent both cardiac computed tomography angiography (CTA) and transesophageal echocardiography (TEE) before left atrial appendage occlusion (LAAO) were included in the study. The patients were allocated into the CR group or non-CR group based on left atrial appendage (LAA) angiography. Patient-specific models were reconstructed to evaluate time-averaged wall shear stress (TAWSS), oscillatory shear index (OSI), relative residence time (RRT), and endothelial cell activation potential (ECAP). Additionally, the incidence of thrombosis was predicted using residence time (RT) at different time-points.

Results

TAWSS was lower [median (Interquartile Range) 0.27 (0.19-0.47) vs 1.35 (0.92-1.79), p < 0.001] in LAA compared to left atrium. In contrast, RRT [1438 (409.70-13869) vs 2.23 (1.81-3.14), p < 0.001] and ECAP [122.70 (30.01-625.70) vs 0.19 (0.16-0.27), p < 0.001)] was higher in the LAA. The patients in the CR group had significantly higher RRT [(mean ± SD) 16274 ± 11797 vs 639.70 ± 595.20, p = 0.009] and ECAP [610.80 ± 365.30 vs 54.26 ± 54.38, p = 0.004] in the LAA compared to the non-CR group. Additionally, patients with CR had a wider range of thrombus-prone regions [0.44(0.27-0.66)% vs 0.05(0.03-0.27)%, p = 0.009] at the end of the 15th cardiac cycle.

Conclusions

These findings suggest that CR might be an indicator of high-risk thrombus formation in the LAA. And CT-based CFD simulation may be a feasible substitute for the evaluation of LAA thrombotic risk in patients with AF, especially in patients with CR.

The role of the pulmonary veins on left atrial flow patterns and thrombus formation

Atrial fibrillation (AF) is the most common human arrhythmia, forming thrombi mostly in the left atrial appendage (LAA). However, the relation between LAA morphology, blood patterns and clot formation is not yet fully understood. Furthermore, the impact of anatomical structures like the pulmonary veins (PVs) have not been thoroughly studied due to data acquisition difficulties. In-silico studies with flow simulations provide a detailed analysis of blood flow patterns under different boundary conditions, but a limited number of cases have been reported in the literature. To address these gaps, we investigated the influence of PVs on LA blood flow patterns and thrombus formation risk through computational fluid dynamics simulations conducted on a sizeable cohort of 130 patients, establishing the largest cohort of patient-specific LA fluid simulations reported to date. The investigation encompassed an in-depth analysis of several parameters, including pulmonary vein orientation (e.g., angles) and configuration (e.g., number), LAA and LA volumes as well as their ratio, flow, and mass-less particles. Our findings highlight the total number of particles within the LAA as a key parameter for distinguishing between the thrombus and non-thrombus groups. Moreover, the angles between the different PVs play an important role to determine the flow going inside the LAA and consequently the risk of thrombus formation. The alignment between the LAA and the main direction of the left superior pulmonary vein, or the position of the right pulmonary vein when it exhibits greater inclination, had an impact to distinguish the control group vs. the thrombus group. These insights shed light on the intricate relationship between PV configuration, LAA morphology, and thrombus formation, underscoring the importance of comprehensive blood flow pattern analyses.

Validated respiratory drug deposition predictions from 2D and 3D medical images with statistical shape models and convolutional neural networks

For the one billion sufferers of respiratory disease, managing their disease with inhalers crucially influences their quality of life. Generic treatment plans could be improved with the aid of computational models that account for patient-specific features such as breathing pattern, lung pathology and morphology. Therefore, we aim to develop and validate an automated computational framework for patient-specific deposition modelling. To that end, an image processing approach is proposed that could produce 3D patient respiratory geometries from 2D chest X-rays and 3D CT images. We evaluated the airway and lung morphology produced by our image processing framework, and assessed deposition compared to in vivo data. The 2D-to-3D image processing reproduces airway diameter to 9% median error compared to ground truth segmentations, but is sensitive to outliers of up to 33% due to lung outline noise. Predicted regional deposition gave 5% median error compared to in vivo measurements. The proposed framework is capable of providing patient-specific deposition measurements for varying treatments, to determine which treatment would best satisfy the needs imposed by each patient (such as disease and lung/airway morphology). Integration of patient-specific modelling into clinical practice as an additional decision-making tool could optimise treatment plans and lower the burden of respiratory diseases.

A comprehensive stroke risk assessment by combining atrial computational fluid dynamics simulations and functional patient data

Stroke, a major global health concern often rooted in cardiac dynamics, demands precise risk evaluation for targeted intervention. Current risk models, like the CHA2DS2-VASc score, often lack the granularity required for personalized predictions. In this study, we present a nuanced and thorough stroke risk assessment by integrating functional insights from cardiac magnetic resonance (CMR) with patient-specific computational fluid dynamics (CFD) simulations. Our cohort, evenly split between control and stroke groups, comprises eight patients. Utilizing CINE CMR, we compute kinematic features, revealing smaller left atrial volumes for stroke patients. The incorporation of patient-specific atrial displacement into our hemodynamic simulations unveils the influence of atrial compliance on the flow fields, emphasizing the importance of LA motion in CFD simulations and challenging the conventional rigid wall assumption in hemodynamics models. Standardizing hemodynamic features with functional metrics enhances the differentiation between stroke and control cases. While standalone assessments provide limited clarity, the synergistic fusion of CMR-derived functional data and patient-informed CFD simulations offers a personalized and mechanistic understanding, distinctly segregating stroke from control cases. Specifically, our investigation reveals a crucial clinical insight: normalizing hemodynamic features based on ejection fraction fails to differentiate between stroke and control patients. Differently, when normalized with stroke volume, a clear and clinically significant distinction emerges and this holds true for both the left atrium and its appendage, providing valuable implications for precise stroke risk assessment in clinical settings. This work introduces a novel framework for seamlessly integrating hemodynamic and functional metrics, laying the groundwork for improved predictive models, and highlighting the significance of motion-informed, personalized risk assessments.

Physics-informed neural networks (PINNs) for 4D hemodynamics prediction: An investigation of optimal framework based on vascular morphology

Hemodynamic parameters are of great significance in the clinical diagnosis and treatment of cardiovascular diseases. However, noninvasive, real-time and accurate acquisition of hemodynamics remains a challenge for current invasive detection and simulation algorithms. Here, we integrate computational fluid dynamics with our customized analysis framework based on a multi-attribute point cloud dataset and physics-informed neural networks (PINNs)-aided deep learning modules. This combination is implemented by our workflow that generates flow field datasets within two types of patient personalized models - aorta with fine coronary branches and abdominal aorta. Deep learning modules with or without an antecedent hierarchical structure model the flow field development and complete the mapping from spatial and temporal dimensions to 4D hemodynamics. 88,000 cases on 4 randomized partitions in 16 controlled trials reveal the hemodynamic landscape of spatio-temporal anisotropy within two types of personalized models, which demonstrates the effectiveness of PINN in predicting the space-time behavior of flow fields and gives the optimal deep learning framework for different blood vessels in terms of balancing the training cost and accuracy dimensions. The proposed framework shows intentional performance in computational cost, accuracy and visualization compared to currently prevalent methods, and has the potential for generalization to model flow fields and corresponding clinical metrics within vessels at different locations. We expect our framework to push the 4D hemodynamic predictions to the real-time level, and in statistically significant fashion, applicable to morphologically variable vessels.

Computational modeling of hemodynamics and risk of thrombosis in the left atrial appendage using patient-specific blood viscosity and boundary conditions at the mitral valve

Hemodynamics play a vital role for the risk of thrombosis in the left atrial appendage (LAA) and left atrium (LA) for patients with atrial fibrillation. Accurate prediction of hemodynamics in the LA can provide important guidance for assessing the risk of thrombosis in the LAA. Patient specificity is a crucial factor in representing the true hemodynamic fields. In this study, we investigated the effects of blood rheology (as a function of hematocrit and shear rate), as well as patient-specific mitral valve (MV) boundary conditions (MV area and velocity profiles measured by ultrasound) on the hemodynamics and thrombosis potential of the LAA. Four scenarios were setup with different degrees of patient specificity. Though using a constant blood viscosity can classify the thrombus and non-thrombus patients for all the hemodynamic indicators, the risk of thrombosis was underestimated for all patients compared with patient-specific viscosities. The results with least patient specificities showed that patients prone to thrombosis predicted by three hemodynamic indicators were inconsistent with clinical observations. Moreover, though patients had the same MV inlet flow rate, different MV models lead to different trends in the risk of thrombosis in different patients. We also found that endothelial cell activation potential and relative residence time can effectively distinguish thrombus and non-thrombus patients for all the scenarios, relatively insensitive to patient specificities. Overall, the findings of this study provide useful insights on patients-specific hemodynamic simulations of the LA.

Latest Developments in Adapting Deep Learning for Assessing TAVR Procedures and Outcomes

Aortic valve defects are among the most prevalent clinical conditions. A severely damaged or non-functioning aortic valve is commonly replaced with a bioprosthetic heart valve (BHV) via the transcatheter aortic valve replacement (TAVR) procedure. Accurate pre-operative planning is crucial for a successful TAVR outcome. Assessment of computational fluid dynamics (CFD), finite element analysis (FEA), and fluid-solid interaction (FSI) analysis offer a solution that has been increasingly utilized to evaluate BHV mechanics and dynamics. However, the high computational costs and the complex operation of computational modeling hinder its application. Recent advancements in the deep learning (DL) domain can offer a real-time surrogate that can render hemodynamic parameters in a few seconds, thus guiding clinicians to select the optimal treatment option. Herein, we provide a comprehensive review of classical computational modeling approaches, medical imaging, and DL approaches for planning and outcome assessment of TAVR. Particularly, we focus on DL approaches in previous studies, highlighting the utilized datasets, deployed DL models, and achieved results. We emphasize the critical challenges and recommend several future directions for innovative researchers to tackle. Finally, an end-to-end smart DL framework is outlined for real-time assessment and recommendation of the best BHV design for TAVR. Ultimately, deploying such a framework in future studies will support clinicians in minimizing risks during TAVR therapy planning and will help in improving patient care.

Increased Blood Residence Time as Markers of High-Risk Patent Foramen Ovale

Previous investigations have suggested that patients with patent foramen ovale (PFO) often have an atrial dysfunction, like to that observed in patients with atrial fibrillation (AF) which may concur to an increased risk of cryptogenic stroke. The aim of the study is to compare the atrial resident time (Rt) of PFO patients to those with sinus rhythm (SR) and AF using patient-specific 3D computational fluid dynamics (CFD) analysis. Models of left atrium (LA) hemodynamics were obtained from time-resolved CT scans and transthoracic echocardiography (TTE). Enrolled patients were divided into three groups: 30 healthy subjects with SR, 30 with PFO, and 30 with AF without PFO. Blood stasis was evaluated by determining the blood residence time (Rt) distribution in the LA and left atrial appendage (LAA). Overall, 90 patients (mean age 47.4 ± 7.5 years, 51 males) were included into the analysis. PFO patients exhibit higher mean Rt values compared to healthy subjects (2.65 ± 0.2 vs 1.5 ± 0.2 s). Conversely, AF patients presented higher Rt when compared to PFO patients (2.9 ± 0.3 vs 2.3 ± 0.2 s). Moreover, PFO patients presenting cerebral lesions at magnetic resonance imaging have a higher Rt compared to those without (2.9 ± 0.3 vs 2.3 ± 0.2 s, respectively, p < 0.001). PFO patients have a higher degree of atrial Rt compared to healthy subjects similar to that observed in AF patients. The higher mean LA Rt values offer an insight into the pathophysiological mechanism linking PFO with cryptogenic stroke and might be a marker of high-risk PFOs.

Machine Learning and the Conundrum of Stroke Risk Prediction

Stroke is a leading cause of death worldwide. With escalating healthcare costs, early non-invasive stroke risk stratification is vital. The current paradigm of stroke risk assessment and mitigation is focused on clinical risk factors and comorbidities. Standard algorithms predict risk using regression-based statistical associations, which, while useful and easy to use, have moderate predictive accuracy. This review summarises recent efforts to deploy machine learning (ML) to predict stroke risk and enrich the understanding of the mechanisms underlying stroke. The surveyed body of literature includes studies comparing ML algorithms with conventional statistical models for predicting cardiovascular disease and, in particular, different stroke subtypes. Another avenue of research explored is ML as a means of enriching multiscale computational modelling, which holds great promise for revealing thrombogenesis mechanisms. Overall, ML offers a new approach to stroke risk stratification that accounts for subtle physiologic variants between patients, potentially leading to more reliable and personalised predictions than standard regression-based statistical associations.

Deep learning-based automatic left atrial appendage filling defects assessment on cardiac computed tomography for clinical and subclinical atrial fibrillation patients

Rationale and objectives

Selecting region of interest (ROI) for left atrial appendage (LAA) filling defects assessment can be time consuming and prone to subjectivity. This study aimed to develop and validate a novel artificial intelligence (AI), deep learning (DL) based framework for automatic filling defects assessment on CT images for clinical and subclinical atrial fibrillation (AF) patients.

Materials and methods

A total of 443,053 CT images were used for DL model development and testing. Images were analyzed by the AI framework and expert cardiologists/radiologists. The LAA segmentation performance was evaluated using Dice coefficient. The agreement between manual and automatic LAA ROI selections was evaluated using intraclass correlation coefficient (ICC) analysis. Receiver operating characteristic (ROC) curve analysis was used to assess filling defects based on the computed LAA to ascending aorta Hounsfield unit (HU) ratios.

Results

A total of 210 patients (Group 1: subclinical AF, n = 105; Group 2: clinical AF with stroke, n = 35; Group 3: AF for catheter ablation, n = 70) were enrolled. The LAA volume segmentation achieved 0.931-0.945 Dice scores. The LAA ROI selection demonstrated excellent agreement (ICC ≥0.895, p < 0.001) with manual selection on the test sets. The automatic framework achieved an excellent AUC score of 0.979 in filling defects assessment. The ROC-derived optimal HU ratio threshold for filling defects detection was 0.561.

Conclusion

The novel AI-based framework could accurately segment the LAA region and select ROIs while effectively avoiding trabeculae for filling defects assessment, achieving close-to-expert performance. This technique may help preemptively detect the potential thromboembolic risk for AF patients.

Finding low CHA2DS2-VASc scores unreliable? Why not give morphological and hemodynamic methods a try?

Patients with atrial fibrillation (AF) suffer from a high risk of thrombosis. Currently, the CHA2DS2-VASc score is the most widely used tool for risk stratification in patients with AF, but it has disappointing accuracy and limited predictive value, especially in those with low scores. Thrombi in patients with AF mostly grow in their left atrial appendages (LAA), which is directly related to the abnormal morphology of the LAA or the left atrium and the unusual hemodynamic state around LAA, which may sensitively evaluate the risk of thrombosis complications in patients with AF and bring bases to clinical plans of medication and operation. Therefore, we investigated the research progress of hemodynamic and morphological studies about the predictive value of thrombosis risk in patients with AF, intending to discuss the prediction potential of morphological and hemodynamic indexes when compared with the presently used CHA2DS2-VASc system and how to build a more precise thromboembolic event prediction model for patients with AF.

Adversarial-based latent space alignment network for left atrial appendage segmentation in transesophageal echocardiography images

Left atrial appendage (LAA) is a leading cause of atrial fibrillation and thrombosis in cardiovascular disease. Clinicians can rely on LAA occlusion (LAAO) to effectively prevent and treat ischaemic strokes attributed to the LAA. The correct selection of the LAAO is one of the most critical stages in the successful surgical process, which relies on the quantification of the anatomical structure of the LAA for successful intervention in LAAO. In this paper, we propose an adversarial-based latent space alignment framework for LAA segmentation in transesophageal echocardiography (TEE) images by introducing prior knowledge from the label. The proposed method consists of an LAA segmentation network, a label reconstruction network, and a latent space alignment loss. To be specific, we first employ ConvNeXt as the backbone of the segmentation and reconstruction network to enhance the feature extraction capability of the encoder. The label reconstruction network then encodes the prior shape features from the LAA labels to the latent space. The latent space alignment loss consists of the adversarial-based alignment and the contrast learning losses. It can motivate the segmentation network to learn the prior shape features of the labels, thus improving the accuracy of LAA edge segmentation. The proposed method was evaluated on a TEE dataset including 1,783 images and the experimental results showed that the proposed method outperformed other state-of-the-art LAA segmentation methods with Dice coefficient, AUC, ACC, G-mean, and Kappa of 0.831, 0.917, 0.989, 0.911, and 0.825, respectively.

Rule-based definition of muscle bundles in patient-specific models of the left atrium

Atrial fibrillation (AF) is the most common arrhythmia encountered clinically, and as the population ages, its prevalence is increasing. Although the CHA2DS2- VASc score is the most used risk-stratification system for stroke risk in AF, it lacks personalization. Patient-specific computer models of the atria can facilitate personalized risk assessment and treatment planning. However, a challenge faced in creating such models is the complexity of the atrial muscle arrangement and its influence on the atrial fiber architecture. This work proposes a semi-automated rule-based algorithm to generate the local fiber orientation in the left atrium (LA). We use the solutions of several harmonic equations to decompose the LA anatomy into subregions. Solution gradients define a two-layer fiber field in each subregion. The robustness of our approach is demonstrated by recreating the fiber orientation on nine models of the LA obtained from AF patients who underwent WATCHMAN device implantation. This cohort of patients encompasses a variety of morphology variants of the left atrium, both in terms of the left atrial appendages (LAAs) and the number of pulmonary veins (PVs). We test the fiber construction algorithm by performing electrophysiology (EP) simulations. Furthermore, this study is the first to compare its results with other rule-based algorithms for the LA fiber architecture definition available in the literature. This analysis suggests that a multi-layer fiber architecture is important to capture complex electrical activation patterns. A notable advantage of our approach is the ability to reconstruct the main LA fiber bundles in a variety of morphologies while solving for a small number of harmonic fields, leading to a comparatively straightforward and reproducible approach.

WSSNet: Aortic Wall Shear Stress Estimation Using Deep Learning on 4D Flow MRI

Wall shear stress (WSS) is an important contributor to vessel wall remodeling and atherosclerosis. However, image-based WSS estimation from 4D Flow MRI underestimates true WSS values, and the accuracy is dependent on spatial resolution, which is limited in 4D Flow MRI. To address this, we present a deep learning algorithm (WSSNet) to estimate WSS trained on aortic computational fluid dynamics (CFD) simulations. The 3D CFD velocity and coordinate point clouds were resampled into a 2D template of 48 × 93 points at two inward distances (randomly varied from 0.3 to 2.0 mm) from the vessel surface (“velocity sheets”). The algorithm was trained on 37 patient-specific geometries and velocity sheets. Results from 6 validation and test cases showed high accuracy against CFD WSS (mean absolute error 0.55 ± 0.60 Pa, relative error 4.34 ± 4.14%, 0.92 ± 0.05 Pearson correlation) and noisy synthetic 4D Flow MRI at 2.4 mm resolution (mean absolute error 0.99 ± 0.91 Pa, relative error 7.13 ± 6.27%, and 0.79 ± 0.10 Pearson correlation). Furthermore, the method was applied on in vivo 4D Flow MRI cases, effectively estimating WSS from standard clinical images. Compared with the existing parabolic fitting method, WSSNet estimates showed 2–3 × higher values, closer to CFD, and a Pearson correlation of 0.68 ± 0.12. This approach, considering both geometric and velocity information from the image, is capable of estimating spatiotemporal WSS with varying image resolution, and is more accurate than existing methods while still preserving the correct WSS pattern distribution.

Sensitivity Analysis of In Silico Fluid Simulations to Predict Thrombus Formation after Left Atrial Appendage Occlusion

Atrial fibrillation (AF) is nowadays the most common human arrhythmia and it is considered a marker of an increased risk of embolic stroke. It is known that 99% of AF-related thrombi are generated in the left atrial appendage (LAA), an anatomical structure located within the left atrium (LA). Left atrial appendage occlusion (LAAO) has become a good alternative for nonvalvular AF patients with contraindications to anticoagulants. However, there is a non-negligible number of device-related thrombus (DRT) events, created next to the device surface. In silico fluid simulations can be a powerful tool to better understand the relation between LA anatomy, haemodynamics, and the process of thrombus formation. Despite the increasing literature in LA fluid modelling, a consensus has not been reached yet in the community on the optimal modelling choices and boundary conditions for generating realistic simulations. In this line, we have performed a sensitivity analysis of several boundary conditions scenarios, varying inlet/outlet and LA wall movement configurations, using patient-specific imaging data of six LAAO patients (three of them with DRT at follow-up). Mesh and cardiac cycle convergence were also analysed. The boundary conditions scenario that better predicted DRT cases had echocardiography-based velocities at the mitral valve outlet, a generic pressure wave from an AF patient at the pulmonary vein inlets, and a dynamic mesh approach for LA wall deformation, emphasizing the need for patient-specific data for realistic simulations. The obtained promising results need to be further validated with larger cohorts, ideally with ground truth data, but they already offer unique insights on thrombogenic risk in the left atria.