You are currently viewing a new version of our website. To view the old version click .
MDPIMDPI
Search all articles
Journal of Clinical Medicine
Download PDF
  • Review
  • Open Access

29 September 2025

Advanced Computer Simulation Based on Cardiac Imaging in Planning of Structural Heart Disease Interventions

,
,
and
1
Department of Medicine, Staten Island University Hospital, New York, NY 10305, USA
2
Division of Cardiology, Columbia University Medical Center, New York, NY 10032, USA
3
Division of Cardiovascular Imaging, St. Francis Hospital and Catholic Health, Roslyn, NY 11576, USA
*
Author to whom correspondence should be addressed.
J. Clin. Med.2025, 14(19), 6885;https://doi.org/10.3390/jcm14196885
This article belongs to the Special Issue Cardiac Imaging: Current Applications and Future Perspectives
Version Notes
Order Reprints

Abstract

The rapid expansion of structural heart interventions over the past decade has created unprecedented challenges in procedural planning and complication prediction. While traditional imaging provides essential anatomical information, translating two-dimensional images into a comprehensive three-dimensional understanding of complex cardiac structures remains challenging. This review encompasses finite element analysis (FEA), computational fluid dynamics (CFD), and fluid–structure interaction (FSI) technologies across major structural heart procedures, including transcatheter aortic valve implantation (TAVI), transcatheter mitral valve interventions, and left atrial appendage occlusion (LAAO). We evaluated the technical foundations, clinical validation studies, and practical applications of various simulation platforms. Advanced computer simulation has demonstrated feasibility and clinical utility across multiple structural heart procedures. Computer simulation for structural heart interventions has evolved from a proof of concept to clinical implementation, with growing evidence of procedural planning benefits in TAVI and LAAO. While feasibility has been established across multiple intervention types, this field requires larger validation studies to demonstrate accuracy and clinical outcome improvements. Future directions include integration of machine learning, real-time simulation capabilities, and expanding applications to complex anatomies and redo procedures. This technology represents an emerging paradigm that may facilitate precision medicine in structural heart interventions, with potential for significant improvements in procedural success and patient safety.

1. Introduction

There has been exponential growth in structural heart disease interventions over the past decade [1,2]. The landmark approval of transcatheter aortic valve implantation (TAVI) with the Edwards SAPIENTM transcatheter heart valve in 2011 led to the rapid progression of this field, with subsequent expansion addressing different pathologies across all cardiac valves and structural components [3,4]. This further led to a proliferation of novel devices, new innovative techniques, and expanding indications, creating a need for clinicians to optimize patient outcomes [5].
There are several factors that compound the complexity of structural heart interventions, including significant anatomical variability among patients and procedural decision making [6]. This differs from traditional open surgical approaches, where visualization and tactile feedback guide interventions. Instead, transcatheter approaches rely heavily on advanced imaging for procedural planning and execution [7]. Subsequently, there has been an evolution in the role of imaging in structural heart interventions beyond initial diagnosis to include screening for treatment selection, device sizing, and prediction of potential complications [8,9].
Pre-procedural planning has progressed to encompass multimodality imaging as the cornerstone of patient assessment [10]. Computed tomography (CT), echocardiography (echo), and cardiac magnetic resonance imaging (CMR) supply data regarding cardiac structure, function, and hemodynamics that inform procedural strategy. However, translating these images into a comprehensive three-dimensional understanding of complex cardiac structures remains challenging [11,12].
To address this challenge, three-dimensional printing has emerged as one approach to create physical representations of patient-specific cardiac anatomy [13]. These models enable hands-on simulation and device testing before the actual procedure, potentially reducing the procedural time and complications [14]. Despite these advantages, 3D printing has significant limitations. The process is costly, time-consuming, and labor-intensive, which restricts its routine clinical application [15]. More importantly, physical models typically represent static anatomical states and fail to capture the dynamic motion of cardiac structures throughout the cardiac cycle [16,17]. Additionally, the material properties of 3D-printed models cannot accurately replicate the biomechanical characteristics of cardiac tissues, which limits their utility in predicting tissue–device interactions [18,19].
Prior reviews of the simulation space have focused on specific procedures [20] or have broadly discussed many subjects, such as AI and 3D printing [17,21,22]. This review adds to recent publications by focusing in depth on the specific topic of advanced computational modeling across the breadth of structural heart interventions. This review discusses computational techniques such as finite element analysis (FEA), computational fluid dynamics (CFD), and fluid–structure interaction (FSI) methodologies, which are specifically applied to imaging-derived cardiac models. The primary focus is a review of imaging-derived advanced computer simulation techniques and the literature within the structural heart disease space.

2. Rationale for Advanced Computer Simulation

These limitations highlight the importance of computational modeling in structural heart interventions. Computer simulation offers several advantages over physical modeling, enabling the creation of patient-specific dynamic models that represent cardiac structures throughout different phases of the cardiac cycle [23]. Virtual devices can be deployed within these models to predict interactions with patient anatomy while accounting for tissue characteristics, flow parameters, regional calcifications, and structural deformation [17]. While basic imaging predictors for many of the typical procedural complications are well known within the literature, their positive and negative predictive values are modest, which suggests interpatient variability and the need for patient-specific modeling. These simulations can theoretically anticipate potential complications, optimize device selection and positioning, and ultimately improve procedural outcomes [20].
The integration of these computer simulation tools into clinical practice has been facilitated by the development of specialized software platforms tailored to specific structural heart interventions [21]. These platforms vary in their capabilities, accessibility, and validation, creating opportunities and challenges for clinicians seeking to implement simulation-based planning in their practices [24].
This paper provides a comprehensive overview of current computer simulation technologies for structural heart interventions, examining their technical foundations, clinical applications, and future directions. Through this analysis, we aim to elucidate the potential for computational modeling to advance the field of structural heart interventions while acknowledging the practical constraints that may influence its adoption in routine clinical practice (Figure 1).
Figure 1. Integrative clinical workflow for imaging-derived simulation in structural heart interventions, from initial imaging and computational modeling to clinical outcomes and continuous refinement.

Clinical Translation and Validation Evidence

Clinical translation of cardiac simulation has demonstrated a measurable impact through several landmark validation studies. The PREDICT-LAA randomized controlled trial (n = 200) represents the first prospective, randomized evidence that AI-enabled computational modeling significantly improves procedural outcomes, showing 40% more complete LAA occlusion (61.1% vs. 44.0%, p = 0.03), a 60% reduction in repositioning devices > three times (10.0% vs. 22.7%, p = 0.02), and 80% fewer device retraction requirements (3.0% vs. 16.5%, p < 0.01) compared to standard planning [25]. Similarly, the PRECISE-TAVI study validated FEA predictions of paravalvular leak (PVL cutoff of 12.2 mL/s, effectively discriminating more than or equal to trace PVL and less than trace PVL with an AUC of 0.69) and conduction disturbances (contact pressure index > 11.5, predicting a permanent pacemaker with an AUC of 0.83) in challenging anatomies [26]. Commercially available platforms are increasingly being developed with the goal of minimizing TAVR-related complications and associated healthcare costs.

3. Types of Computer Simulations and Platforms

3.1. Basic Computer Modeling

Basic computer modeling focuses primarily on anatomical visualization and measurement, providing fundamental insights for device sizing and procedural planning [27]. Basic computer modeling has existed since the early days of TAVI and consists of an embedded geometrical shape inserted into a cardiac computed tomography angiography (CCTA) image. This type of modeling can be used to determine the basic fit of a transcatheter valve in the target zone. A simple example of this is modeling a cylinder of the predicted THV size for TAVI. Examples of vendors in this space include 3mensioTM, Circle Cardiovascular Imaging, and Laralab (Figure 2).
Figure 2. Basic segmentation and modeling of the aortic root. Red panel: The aortic root was auto-segmented, and basic measurements of the aortic annulus dimensions were derived using Laralab’s heart.ai software (Munich, Germany, https://www.laralab.com/, accessed on 15 July 2025). Blue panel: Automated segmentation was performed by detection of each aortic valve leaflet nadir, and aortic annulus measurements were performed using 3mensioTM (Pie Medical Imaging, Maastricht, The Netherlands).

3.2. Advanced Computer Simulation

Advanced simulation typically involves the creation of a digital twin of the structure(s) in question and use of finite element analysis (FEA) to predict tissue deformation and responses to interventions. There is research on individual centers or labs predicting implant success. In the past several years, vendors dedicated to advanced simulation have emerged. Examples of technologies in this space include FeOPS Heart GuideTM and DASI simulations. Some technologies also model flow, which typically involves performing computational fluid dynamics (CFD) and fluid–structure interaction (FSI) analyses in a gap region or area from the FEA. Examples of available platforms include FEOPSTM, DASI, VIDAA, and AnsysTM. This review will focus primarily on advanced computer simulation.

3.3. Platform–Device Compatibility

Current simulation platforms demonstrate varying capabilities across device families and procedures, spanning TAVI, LAAO, and emerging TMVR applications. This is largely dependent on collaborations with device manufacturers, who can provide key material properties and device designs in order to facilitate computer simulations. Without adequate info, computational simulation platforms become less accurate. Additionally, given the workload for computational heavy lifting, the platforms tend to focus on a limited number of procedures and thus self-limit their portfolios of devices (Table 1).
Table 1. Key clinical validation evidence for cardiac simulation platforms.

4. Computer Simulation for Specific Structural Heart Interventions

Compared to traditional surgical repairs of valves, transcatheter valvular interventions provide minimally invasive alternatives to patients with valvular heart disease. Transcatheter procedures encompass unique risks compared to open heart surgery. Computer simulation models attempt to mitigate these risks by ensuring procedural planning that is specific to patients, assisting in optimal device selection, and predicting complications (Figure 3).
Figure 3. Computer simulation of TAVI and mitral valve interventions. Upper left: Simulation of tissue interaction for an S3 in an Evolut redo TAVI. Upper right: A 3D mesh with simulation of anterior mitral leaflet laceration using electrosurgery. Lower left: Simulation of a valve in an MAC procedure, predicting residual paravalvular leak in the mitral commissure (red arrow). Lower right: An actual post-procedural 3D color Doppler TEE image demonstrating paravalvular leak in the same commissure that was predicted. The images were provided by DASI simulations, Dublin, Ohio.

4.1. TAVI

As TAVI was the first approved transcatheter valve procedure, the majority of the advanced computer simulation literature centers around it.

4.1.1. Optimal Sizing

The major TAVI vendors have well-developed sizing charts to address different ranges. Basic computer simulation can be applied to cases to visualize a static valve within the segmented aortic root. In current practice, this is the workhorse method for most TAVI programs.

4.1.2. Anatomical Rupture

Aortic root rupture (ARR) is a rare but serious complication of TAVI with a mortality rate that has been reported to be as high as 48% [29]. This is in a continuum with paravalvular leak (PVL), as overexpansion of a balloon or valve in the presence of severe calcification may lead to root rupture, whereas underexpansion can lead to PVL. The difficulty in predicting ARR is evidenced by the dearth of research on the subject. In 2013, Barbanti et al. stated that ≥20% prosthesis oversizing of a balloon-expandable valve and moderate/severe LVOT calcifications were predictors of ARR [29]. In 2017, Girdauskas et al. described six patients who had ruptured their LVOT in the anatomically weaker area between the left fibrous trigone and the left–right commissure within the muscular LVOT [30]. In theory, computer simulation would benefit prediction of the rare but morbid complication of ARR. There is only one computer simulation study in the literature. It was performed by Wang et al., who employed finite element analysis on one retrospective ARR case and two prospective cases [31]. The results were mixed, again highlighting the real-world difficulties in understanding this complex phenomenon.

4.1.3. Paravalvular Leak

Aortic post-TAVI PVL has become less common over time, particularly PVL of moderate or greater severity. However, mild PVL is common, and some analyses have demonstrated increased morbidity and/or mortality, even at mild levels [32]. Basic predictors are well known, including transcatheter heart valve (THV) undersizing, the THV type (it is more common in self-expanding valves), significant calcifications, valve positioning, and a native bicuspid valve. However, there is significant interpatient variability. Many retrospective, observational studies have shown the ability to predict PVL retrospectively from computer simulations [33,34,35,36,37]. Mao et al., Bianchi et al., and Ghosh et al. demonstrated the feasibility of using FEA or FEA-FSI modeling to predict PVL based on the orientations and deployment depths of THVs [38]. Luraghi et al. and Nappi et al. demonstrated the effects of calcium on PVL formation [35,37]. Hokken et al. demonstrated an AUC of 0.69 for prediction of PVL after Evolut TAVI, with a simulated flow rate cutoff of 12.2 mL/s to predict > trace PVL [26].

4.1.4. Percutaneous Alternatives for Coronary Artery Obstruction

Coronary obstruction (CO) is yet another rare but morbid complication of TAVI. Known anatomical risk factors for native TAVI based on basic modeling studies include a low coronary height of <12 mm and a low sinus of Valsalva diameter of <30 mm [39], and a low valve-to-coronary (VTC) distance of <4 mm [40]. For valve-in-valve (VIV) TAVI, known anatomical risk factors include a VTC of <4 mm and externally sewn bioprosthetic surgical valve leaflets. There has been very little work using advanced computer simulation to predict the risk of coronary obstruction. Heitkemper et al. studied a retrospective cohort of TAVI patients and used FEA to demonstrate that the VTC distance indexed to the coronary diameter was more predictive of CO than the typical metrics of coronary height and sinus of Valsalva diameter [41]. Holst et al. recently reported the feasibility of prospective use of CO FEA computer modeling prior to TAVI cases. In 29 TAVI cases with high predicted CO risk, 21 underwent a procedure or protective maneuver to avoid coronary obstruction, and none of the patients experienced CO [42].

4.1.5. Leaflet Thrombosis

Leaflet thrombosis is a common sequela of TAVI, with a reported prevalence as high as 30–40%. Leaflet thrombosis appears to be more common in THVs than in surgical bioprosthetic valves. Known anatomical risk factors include a native bicuspid valve, moderate or severe PVL, a large anatomy, and the THV type. Attempts to predict leaflet thrombosis to date have focused on CFD analysis of stagnation of blood and/or neosinus washout [38,43,44,45]. Reduced neosinus washout and blood stagnation have been associated with thrombus formation in computational studies. Extensive clinical validations have not been performed.

4.1.6. Prediction of Conduction Disturbances

In an initial FEA study by Rocatello et al., the metric of the contact pressure index (CPI), defined as the percentage of the area where contact between the prosthesis and the aorta generate a pressure exceeding 0.1 Mpa, predicted conduction disturbance at a threshold of 14%. However, only 26 patients in the 112-patient cohort were determined to have accurate predictions [46]. In a subsequent cohort of 56 patients with bicuspid or tricuspid aortic valve stenosis that were studied by the same group, the CPI was re-calibrated, which resulted in improved accuracy [47]. In a univariate analysis, only the CPI predicted a new-onset conduction disturbance. The traditional metrics of the oversizing ratio, implantation depth, and difference between the membranous septum length and the implantation depth were not effective for this prediction. The CPI exhibited a higher AUC (0.804) when predicting a new-onset conduction disturbance than the traditional factors. The PRECISE-TAVI trial, a prospective, observational study of computational simulation of complex aortic valve anatomies, similarly demonstrated an AUC of 0.83 for prediction of permanent pacemaker implantation after Evolut-PRO implantation, with a CPI cutoff of 11.5 [26].

4.1.7. TAVI in Bicuspid Aortic Valves

Bicuspid valve TAVI is a complex area. Current THVs are not designed specifically to treat bicuspid aortic valves. Retrospective validation of FEA and CFD simulations has been performed by several groups in small studies [48]. The PRECISE-TAVI trial prospectively studied a group of patients, 17 (22.1%) of which had a bicuspid valve morphology [26]. As mentioned previously, this trial demonstrated the predictive abilities of FEA for permanent pacemaker implantation and PVL. FEA prompted a change in procedural strategy in 35% of the patients in this study. Due to small numbers, the bicuspid aortic valve group was not studied separately.

4.2. TMVR

4.2.1. Mitral Valve-in-Valve TAVI

Given the excellent outcomes for mitral VIV TAVI with a balloon-expandable THV designed for TAVI, basic simulation appears to be adequate for the majority of cases. The sealing zone of a failed surgical bioprosthesis provides an excellent landing zone, and the morbid complication of left ventricular outflow tract obstruction (LVOTO) is rare in comparison to other forms of TMVR.

4.2.2. Mitral Valve in Ring and Valve in Mitral Annular Calcification

A mitral valve in a failed annuloplasty ring (ViR) and a valve in a native mitral annular calcification (ViMAC) are more complex forms of TMVR associated with less favorable procedural outcomes, such as LVOTO and higher mortality rates. These are currently treated using a balloon-expandable THV designed for TAVI. This suggests the need for advanced computer simulation as opposed to valve-in-valve TAVI. Although the feasibility of computer simulation was shown in early examples [49,50,51], progress in this field is limited and data are sparse.

4.2.3. Native TMVR

Dedicated (designed for the mitral valve) TMVR devices for native mitral valve disease (mitral regurgitation without MAC) are mostly in the investigational space, with early CE Mark and FDA approvals. There are no available data on advanced computer simulation in this category.

4.2.4. Mitral Edge-to-Edge Repair

The feasibility of FEA and FSI analysis of the mitral valve leaflet structure and mechanics has been shown by several groups [52,53]. Dabiri et al. developed a framework not only to simulate mitral valve geometry and flow dynamics but also to predict the results of mitral transcatheter edge-to-edge repair (m-TEER) [53]. This group used actual 3D echocardiographic datasets and computational data from the literature to model mitral geometries and simulate the results of m-TEER procedures, including MR and leaflet stress, with single and multiple m-TEER devices. Dabiri et al. demonstrated via FEA that different locations of m-TEER reduced MR and the mitral valve area by different degrees. They emphasized that machine learning (ML) has fewer limitations than FEA overall but requires much larger datasets for analysis. In a separate publication, Dabiri et al. demonstrated that an ML model for m-TEER simulation reduced the computational time from 6 h to less than 1 s [54].
Although there are limited real-world validation data, several proof-of-concept computer simulation studies have been performed. Prescott et al. performed an FEA simulation to confirm that m-TEER could treat MR and prevent posterior leaflet stress and chordal rupture in a hypertrophic cardiomyopathy patient [55]. Hart et al. retrospectively studied the accuracy of pre- and post-m-TEER 3D TEE datasets from two real patients compared to FEA simulations. The FEA simulations only differed by 0.7–0.9 mm in the locations of contours compared to the actual pre-m-TEER datasets [56]. In both patients, FEA simulation accurately predicted the severity and location of post-m-TEER MR. Messika-Zeitoun et al. published the largest simulation analysis for m-TEER computational FEA modeling to date [57]. These authors studied five patients who underwent m-TEER. There was good agreement between the actual post-procedural MR grade and the simulated regurgitant orifice area. However, the agreement between the post-procedural mitral valve area modeling and the mitral valve area measured using 3D TEE was poor, with only one of five patient simulations being accurate.

4.2.5. Limitations in Transcatheter Edge-to-Edge Repair Simulation

Current TEER simulations are in their infancy. For the most part, straightforward and simple anatomies are the current focus. Even with the focus on simple anatomies, predictive accuracy has been mixed, as described above. Non-central and commissural TEER were found to be very prevalent, representing between 38 and 43% of the published data [58,59].
Finite element models experience bending difficulties near the commissures and cannot adequately simulate the complex dynamic interactions between commissural chordae and leaflet tissue during the cardiac cycle [60]. These limitations result in poor prediction accuracy for device trajectory correction and an inability to reliably predict single-leaflet device attachment risk during commissural repairs [60].

4.3. Left Atrial Appendage Occlusion

Data from the NCDR LAAO (left atrial appendage closure) registry has documented excellent procedural outcomes with commercially available LAAO devices in the United States, with an implantation success rate of >98%. Yet one in four patients has evidence of a peri-device leak after 45 days. Recent evidence suggests that even small PDLs (<5 mm) are associated with a higher rate of thromboembolic events; this has therefore further fueled interest in improving procedural planning to achieve complete LAAO [61].
LAA anatomy is complex, and individual anatomy is variable. CT-based computational simulation algorithms allow clinicians to depict device implantation and assess tissue–device interaction as a means to improve procedural efficacy and predict the feasibility of LAAO, peri-device gaps, and the correct device size. Computational fluid simulation models have also been used to predict device-related thrombosis (DRT).

4.3.1. Methods of Analysis for CT-Based Computational Modeling of LAAO

FEops HEARTguideTM (FEops Ghent, Belgium).
FEops HEARTguideTM is a CE-marked and commercially available CT-based simulation technology that gives physicians insight into tissue–device interaction in patient-specific anatomy. Details of how the computational model was developed were previously published [62]. Briefly, computer-generated LAAO devices are virtually implanted into patient-specific anatomy using FEA computational simulation (Abaqus/Explicit finite element solver v170.0, Dassault Systems, Paris, France). Multiple sizes are simulated in different positions (proximal and distal), and frame deformation and device apposition are modeled. In addition, device diameters are assessed to evaluate compression. It is up to the provider to assess the simulations and select the most appropriate device size and position (Figure 4). The FEopsTM simulation technology has been validated for LAAO [50,63].
Figure 4. An example of the FEopsTM patient-specific computational model.
Multiple small to medium-sized prospective and retrospective studies have analyzed the utilization of FEopsTM for LAAO procedural planning (Table 2). The largest of these studies was the PREDICT—LAA study, which looked specifically at the value of FEopsTM for planning LAAO with the Amplatzer AmuletTM device [25]. This was the first and, to date, only prospective, randomized trial of advanced computer simulation for device implantation. Though this study did not meet its primary endpoint (a composite of incomplete LAA closure with a residual grade III to IV distal leak and/or presence of device-related thrombosis on a post-procedural CT scan), there was a 31% decrease in the primary outcome in the FEopsTM group (28.9% in the FEopsTM group versus 41.8% in the standard treatment group, p = 0.08) and thus a trend in favor of CT-based simulation planning. Notably, the trial reached statistical significance (p < 0.05) for the secondary endpoints that focused on procedural efficacy (achievement of complete LAA closure, the number of devices used, the number of device repositionings, the procedural time, the radiation time, and the dye volume). Thus, overall, the trial demonstrated that the FEopsTM computational model offers improvements in procedural safety and efficacy for LAAO with the AmuletTM device [25]. A few of the criticisms of this trial have been the lack of standardization in the control arm and the need for reproduction of the findings with a larger sample. Additionally, these findings cannot be extrapolated to other LAAO devices.
Table 2. Key findings of studies that have evaluated CT-based computational modeling for LAAO.

4.3.2. Virtual Implantation and Device Selection in Left Atrial Appendages (VIDAA) Platform (Universitat Pompeu Fabra, Barcelona, Spain)

The VIDAA platform is a web-based 3D interactive virtual implantation application that allows a clinician to evaluate the LAA and manipulate LAAO device configurations to determine the optimal device selection. An individual’s anatomy is segmented based on a CT scan using 3D slicer and is used to create a 3D surface mesh. Both the CT scan and the surface mesh are then imported into the VIDAA program. CFD simulations are also performed. Two different LAAO devices are available in the platform: WatchmanTM and AmuletTM. The program proposes a set of appropriate LAA devices for a given geometry for the clinician to review. A catheter model is also available in the platform to simulate device delivery [67].

4.3.3. Computational Fluid Dynamics to Predict Device-Related Thrombosis

CFD was first evaluated as a way to stratify the risk of LAA thrombus formation based on LAA morphology in atrial fibrillation patients. A number of studies have demonstrated that geometric characteristics of the LAA play a central role in defining thromboembolic risk [68,69,70,71]. Bosi et al. specifically looked at four different LAA morphologies (chicken wing, cactus, windsock, and cauliflower). First, CT scans were processed using Mimics (Materialise NV, Ghent, Belgium) to obtain the 3D anatomical shape of the LAA and LA. All 3D models were meshed and analyzed using AnsysTM (AnsysTM, Inc., Canonsburg, Pennsylvania). CFD was evaluated in normal physiologic flow and pathological atrial fibrillation conditions. Blood flow patterns were characterized by their velocities and shear strain rates (SSRs). The velocity and SSR decreased from the ostium to the tip of the LAA for all LAA morphologies. In atrial fibrillation, the lowest velocity was observed with the cauliflower morphology and the highest was observed with the windsock morphology. Furthermore, in atrial fibrillation, the cauliflower morphology also showed the lowest SSR [69]. Thus, this study indicated that in pathologic conditions the cauliflower morphology is more prone to thrombus development. The cauliflower morphology has high thrombotic potential, as do morphologies that have multiple small lobes, which was confirmed in other studies [71,72].
Subsequently, CFD has been studied as a way to predict the risk of device-related thrombosis following LAAO. The most common measure evaluated is the endothelial cell activation potential (ECAP) index, which is an in silico thrombosis risk index. The ECAP combines the time-averaged wall shear stress with the oscillatory shear index. In general, low velocities (low wall shear stress values) and complex blood flow patterns (higher oscillatory shear index values) produce higher ECAP indices, which are thought to confer a higher risk of thrombus formation.
Table 3 details studies that have analyzed CFD for predicting DRT. Overall velocities < 0.2 m/s and an ECAP index > 0.5 Pa−1 have been suggested to be predictors of DRT. Another risk factor for DRT that emerged out of these studies is a low velocity along the pulmonary ridge if uncovered, especially if the pulmonary ridge flow cannot be washed out [73].
Table 3. Key findings of studies that have evaluated CFD after LAAO.
In the future, CFD has the potential to help individualize anti-thrombotic regimens post-LAAO. However, there are notable limitations to the CFD methodology that need to be addressed. There is no clear consensus on the optimal boundary conditions to model LA hemodynamics, and this needs to be standardized. This is especially important, as CFD has a high sensitivity to numerical assumptions. There is also a lack of model verification [78]. Lastly, this methodology has been evaluated in small studies and needs to be applied on a larger scale to further our understanding.

4.3.4. Artificial Intelligence (AI) Integration and Enhanced Simulation

Integration of artificial intelligence with cardiac simulation accelerated significantly in 2023–2024. AI applications now span the entire cardiac simulation workflow from automated cardiac structure segmentation to predictive outcome modeling.
AI-powered clinical decision support systems integrate knowledge from basic biology to current pathophysiologic insights and drive real-time image analysis, segmentation, and annotation to guide operators in selecting the most appropriate strategies for individual patients [79]. Technical advances in cardiac simulation AI include physics-informed neural networks (PINNs) that incorporate cardiac biomechanics constraints, reducing CFD computation times while maintaining accuracy within traditional methods. AI-enhanced segmentation using U-Net architectures and automated mesh generation eliminate traditional post-processing artifacts while reducing processing times from hours to minutes [80].
However, significant challenges persist in AI implementation. Data quality limitations affect generalizability, with only 3.6% of FDA-approved AI devices reporting race/ethnicity data and 99.1% lacking socioeconomic information [81]. Deep learning models often operate as “black boxes” with limited interpretability, hindering clinical acceptance [82]. Technical challenges include the need for large, diverse training datasets and patient-specific anatomical variations that challenge model generalization [83]. Current solutions include Gradient-weighted Class Activation Mapping (Grad-CAM) for visualization and Shapley Additive Explanations (SHAP) for feature importance analysis, but standardized frameworks for AI model development and validation are lacking [84].
Current solutions and validation approaches include development of explainable AI frameworks specifically for cardiac simulation, using attention mapping to highlight anatomical features driving device sizing recommendations. Federated learning approaches allow model training across multiple institutions while maintaining patient privacy. However, standardized benchmarks for cardiac simulation AI performance are lacking, with validation studies typically involving single-center cohorts with relatively small samples of less than 200 patients.

4.3.5. Implementation Barriers and Clinical Adoption Challenges

Despite technical advances, significant barriers limit widespread clinical adoption of cardiac simulation technologies. Cost barriers include high infrastructure requirements, including the costs of simulation centers, per-case costs, and substantial investments in specialized personnel and maintenance [85]. Additional expenses include specialized workstation hardware, high-resolution display systems, and training for dedicated technical personnel. Learning curve challenges persist. A survey conducted in 2024 found that 71% of interventional cardiology professionals reported inadequate training related to simulation-based planning, with most training experiences lasting under one week, which is insufficient for skill mastery [86].
Standardization issues represent a critical gap in cardiac simulation implementation. Currently, no standardized protocols exist for CT acquisition parameters optimized for simulation; mesh generation quality control; or simulation result interpretation involving data variability, protocol inconsistency, a lack of external validation, fragmented regulatory requirements, and limited interoperability among software systems [87,88,89].
Technical limitations include computational demands (CFD modeling requires many hours per case, increases memory requirements, and creates processing limitations, preventing real-time intraoperative guidance), moderate overall fidelity compared to actual clinical scenarios, and device-specific compatibility constraints [90,91].
Device and anatomy generalizability remain constrained. These limitations arise from platform-specific capabilities, with most simulation tools optimized for specific anatomical presentations and device families rather than comprehensive coverage of anatomical variations and pathological presentations [89,92], such as bicuspid valves, heavily calcified annuli, and prior surgical interventions, which often fall outside validated simulation parameters. This limits clinical applicability in complex cases, where simulation guidance would be most valuable.

5. Future Directions

Overall, this field needs more robust trials to prove the accuracy of advanced computational modeling for procedural simulation and, most importantly, translation to improvements in procedural and clinical outcomes. More computational advancement and power are needed to make simulations more efficient and allow timely intervention.

5.1. TAVI

As procedures become more complex, the need for advanced computational simulation will increase. The feasibility of redo TAVI (TAV in TAV) procedural simulation has been shown. However, more validation is needed (Figure 5). Lifetime management of aortic stenosis treatment (predicting results of sequential procedures) is another important area of need.
Figure 5. The feasibility of redo TAVI computer simulation. Top panel: Simulation of an S3 in an Evolut valve combination for redo TAVI at different implantation heights. Bottom panel: Simulation of a redo TAVI S3 in an Evolut combination, demonstrating the proximity to the aorta wall, leaflet overhang, and commissural alignment. The images were provided by FEops HeartGuideTM, Gent, Belgium.

5.2. TMVR

Sizing is a concern in some valve-in-ring cases, particularly valve-in-MAC cases, and requires more accurate modeling. The displacement and behavior of MAC can be unpredictable with percutaneous valve implantation. However, the largest concern and most common exclusion for TMVR cases is LVOT obstruction. There is a great need for accurate computational modeling of LVOT obstruction, including the neo-LVOT area, tissue interactions, and displacement of the anterior mitral leaflet.

5.3. Tricuspid Interventions

Efforts are preliminary, but feasibility studies of computer simulation of tricuspid leaflet geometry have been performed. More work will be needed to validate procedural computational simulation as the field matures.

5.4. LAAO

While this is the only intervention that has a randomized, controlled trial as evidence, the results were only valid for a single device type. Computer simulation of other device types needs more validation.

6. Conclusions

The field of computational simulation for structural heart device implantation procedures has made significant progress. Feasibility has been proven. Future impacts in this field will be predicated upon validation of accuracy and improvements in outcomes.

Author Contributions

Conceptualization, A.A. and O.K.; methodology, O.K.; formal analysis, A.A., L.R., and T.V.; data curation, A.A., L.R., and T.V.; writing—original draft preparation, A.A., L.R., and T.V.; writing—review and editing, O.K. and A.A.; visualization, O.K.; supervision, O.K. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Data Availability Statement

No new data were generated or analyzed in support of this research.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
CFDComputational fluid dynamics
FSIFluid–structure interaction
ECAPEndothelial cell activation potential
VIDAAVirtual implantation and device selection in left atrial appendages
ARRAortic root rupture
FEAFinite element analysis
FEopsTMFinite element-optimized patient-specific
DASIDirect analytical surgical individualization
ViRValve in ring
THVTranscatheter heart valve
COCoronary obstruction
ViVValve in valve
ViMACValve in mitral annular calcification
LAAOLeft atrial appendage occlusion
CPIContact pressure index
PDLPeri-device leak
DRTDevice-related thrombosis
LVOTLeft ventricular outflow tract
SSRShear strain rate
MACMitral annular calcification
m-TEERMitral transcatheter edge-to-edge repair
PVLParavalvular leak
TAVITranscatheter aortic valve implantation
TAVRTranscatheter aortic valve replacement

References

  1. Faza, N.N.; Özden Tok, Ö.; Hahn, R.T. Imaging in Structural Heart Disease: The Evolution of a New Subspecialty. JACC Case Rep. 2019, 1, 440–445. [Google Scholar] [CrossRef]
  2. Goel, S.S. The Current and Future Landscape of Structural Heart Interventions. Methodist. Debakey Cardiovasc. J. 2023, 19, 1–3. [Google Scholar] [CrossRef] [PubMed]
  3. Webb, J.G.; Wood, D.A. Current status of transcatheter aortic valve replacement. J. Am. Coll. Cardiol. 2012, 60, 483–492. [Google Scholar] [CrossRef]
  4. Holoshitz, N.; Kavinsky, C.J.; Hijazi, Z.M. The Edwards SAPIEN Transcatheter Heart Valve for Calcific Aortic Stenosis: A Review of the Valve, Procedure, and Current Literature. Cardiol. Ther. 2012, 1, 6. [Google Scholar] [CrossRef][Green Version]
  5. Wunderlich, N.C.; Küx, H.; Kreidel, F.; Birkemeyer, R.; Siegel, R.J.; Darmstadt, D.C.C.; Ulm, U.H. The Changing Paradigm in the Treatment of Structural Heart Disease and the Need for the Interventional Imaging Specialist. Interv. Cardiol. 2016, 11, 135–139. [Google Scholar] [CrossRef][Green Version]
  6. Perpetua, E.M.; Guibone, K.A.; Keegan, P.A.; Palmer, R.; Speight, M.K.; Jagnic, K.; Michaels, J.; Nguyen, R.A.; Pickett, E.S.; Ramsey, D.; et al. Best Practice Recommendations for Optimizing Care in Structural Heart Programs: Planning Efficient and Resource Leveraging Systems (PEARLS). Struct. Heart 2021, 5, 168–179. [Google Scholar] [CrossRef]
  7. Ye, J.; Soon, J.L.; Webb, J. Aortic valve replacement vs. transcatheter aortic valve implantation: Patient selection. Ann. Cardiothorac. Surg. 2012, 1, 194–199. [Google Scholar] [CrossRef]
  8. Corrigan, F.E., 3rd; Gleason, P.T.; Condado, J.F.; Lisko, J.C.; Chen, J.H.; Kamioka, N.; Keegan, P.; Howell, S.; Clements, S.D.; Babaliaros, V.C.; et al. Imaging for Predicting, Detecting, and Managing Complications After Transcatheter Aortic Valve Replacement. JACC Cardiovasc. Imaging 2019, 12, 904–920. [Google Scholar] [CrossRef]
  9. Mooney, J.; Sellers, S.L.; Ohana, M.; Cavalcante, J.; Arepalli, C.; Grover, R.; Kim, U.; Selvakumar, K.; Blanke, P.; Leipsic, J. Imaging for structural heart procedures: Focus on computed tomography. EuroIntervention 2017, 13, AA85–AA96. [Google Scholar] [CrossRef] [PubMed]
  10. Grigoryan, K.; Demetrescu, C.; Kasouridis, I.; Abiola, O.; Masci, P.G.; Oguz, D.; Benedetti, G.; Mak, S.M.; Parwani, P.; Preston, R.; et al. Multimodality Imaging in Valvular Structural Interventions. Card. Fail. Rev. 2022, 8, e31. [Google Scholar] [CrossRef] [PubMed]
  11. Agricola, E.; Ingallina, G.; Ancona, F.; Biondi, F.; Margonato, D.; Barki, M.; Tavernese, A.; Belli, M.; Stella, S. Evolution of interventional imaging in structural heart disease. Eur. Heart J. Suppl. 2023, 25 (Suppl. C), C189–C199. [Google Scholar] [CrossRef]
  12. Vrijkorte, M.G.H.; Khalique, O.K.; Lerakis, S.; Swaans, M.J. Editorial: Imaging in structural heart interventions. Front. Cardiovasc. Med. 2024, 11, 1364243. [Google Scholar] [CrossRef]
  13. Harb, S.C.; Rodriguez, L.L.; Vukicevic, M.; Kapadia, S.R.; Little, S.H. Three-Dimensional Printing Applications in Percutaneous Structural Heart Interventions. Circ. Cardiovasc. Imaging 2019, 12, e009014. [Google Scholar] [CrossRef]
  14. Gómez-Ciriza, G.; Gómez-Cía, T.; Rivas-González, J.A.; Forte, M.N.V.; Valverde, I. Affordable Three-Dimensional Printed Heart Models. Front. Cardiovasc. Med. 2021, 8, 642011. [Google Scholar] [CrossRef] [PubMed]
  15. Lau, I.; Sun, Z. Three-dimensional printing in congenital heart disease: A systematic review. J. Med. Radiat. Sci. 2018, 65, 226–236. [Google Scholar] [CrossRef] [PubMed]
  16. Spanaki, A.; Kabir, S.; Stephenson, N.; van Poppel, M.P.M.; Benetti, V.; Simpson, J. 3D Approaches in Complex CHD: Where Are We? Funny Printing and Beautiful Images, or a Useful Tool? J. Cardiovasc. Dev. Dis. 2022, 9, 269. [Google Scholar] [CrossRef]
  17. Faza, N.N.; Harb, S.C.; Wang, D.D.; Dorpel, M.M.v.D.; Van Mieghem, N.; Little, S.H. Physical and Computational Modeling for Transcatheter Structural Heart Interventions. JACC Cardiovasc. Imaging 2024, 17, 428–440. [Google Scholar] [CrossRef] [PubMed]
  18. Vukicevic, M.; Mosadegh, B.; Min, J.K.; Little, S.H. Cardiac 3D Printing and its Future Directions. JACC Cardiovasc. Imaging 2017, 10, 171–184. [Google Scholar] [CrossRef]
  19. Bhandari, S.; Yadav, V.; Ishaq, A.; Sanipini, S.; Ekhator, C.; Khleif, R.; Beheshtaein, A.; Jhajj, L.K.; Khan, A.W.; Al Khalifa, A.; et al. Trends and Challenges in the Development of 3D-Printed Heart Valves and Other Cardiac Implants: A Review of Current Advances. Cureus 2023, 15, e43204. [Google Scholar] [CrossRef]
  20. Wong, P.; Wisneski, A.D.; Sandhu, A.; Wang, Z.; Mahadevan, V.S.; Nguyen, T.C.; Guccione, J.M. Looking towards the future: Patient-specific computational modeling to optimize outcomes for transcatheter mitral valve repair. Front. Cardiovasc. Med. 2023, 10, 1140379. [Google Scholar] [CrossRef]
  21. Samant, S.; Bakhos, J.J.; Wu, W.; Zhao, S.; Kassab, G.S.; Khan, B.; Panagopoulos, A.; Makadia, J.; Oguz, U.M.; Banga, A.; et al. Artificial Intelligence, Computational Simulations, and Extended Reality in Cardiovascular Interventions. JACC Cardiovasc. Interv. 2023, 16, 2479–2497. [Google Scholar] [CrossRef] [PubMed]
  22. Ooms, J.F.; Wang, D.D.; Rajani, R.; Redwood, S.; Little, S.H.; Chuang, M.L.; Popma, J.J.; Dahle, G.; Pfeiffer, M.; Kanda, B.; et al. Computed Tomography-Derived 3D Modeling to Guide Sizing and Planning of Transcatheter Mitral Valve Interventions. JACC Cardiovasc. Imaging 2021, 14, 1644–1658. [Google Scholar] [CrossRef]
  23. Alkhouli, M.; Hatoum, H.; Piazza, N. Computational Modeling to Guide Structural Heart Interventions: Measure Twice (or Thrice) But Cut Once. JACC Cardiovasc. Interv. 2023, 16, 667–669. [Google Scholar] [CrossRef]
  24. Krishnaswamy, A.; Kassab, J.; Harb, S.C. Beyond Simple Visualization: A New Reality for Structural Heart Interventions? J. Am. Heart Assoc. 2024, 13, e036238. [Google Scholar] [CrossRef]
  25. De Backer, O.; Iriart, X.; Kefer, J.; Nielsen-Kudsk, J.E.; Aminian, A.; Rosseel, L.; Kofoed, K.F.; Odenstedt, J.; Berti, S.; Saw, J.; et al. Impact of Computational Modeling on Transcatheter Left Atrial Appendage Closure Efficiency and Outcomes. JACC Cardiovasc. Interv. 2023, 16, 655–666. [Google Scholar] [CrossRef]
  26. Hokken, T.W.; Wienemann, H.; Dargan, J.; van Ginkel, D.-J.; Dowling, C.; Unbehaun, A.; Bosmans, J.; Bader-Wolfe, A.; Gooley, D.; Swaans, M.; et al. Clinical value of CT-derived simulations of transcatheter-aortic-valve-implantation in challenging anatomies the PRECISE-TAVI trial. Catheter. Cardiovasc. Interv. 2023, 102, 1140–1148. [Google Scholar] [CrossRef]
  27. Raphael, C.E.; Alkhouli, M.; Maor, E.; van Ginkel, D.; Dowling, C.; Unbehaun, A.; Bosmans, J.; Bader-Wolfe, A.; Gooley, R.; Swaans, M.; et al. Building Blocks of Structural Intervention: A Novel Modular Paradigm for Procedural Training. Circ. Cardiovasc. Interv. 2017, 10, e005686. [Google Scholar] [CrossRef] [PubMed]
  28. De Cock, E.; Lochy, S.; Rivero-Ayerza, M.; Lempereur, M.; Cornelis, K.; Debonnaire, P.; Vermeersch, P.; Christiaen, E.; Buysschaert, I. Clinical Value of CT-Based 3D Computational Modeling in Left Atrial Appendage Occlusion: An In-Depth Analysis of the PRECISE LAAO Study. Catheter. Cardiovasc. Interv. 2025, 105, 1356–1364. [Google Scholar] [CrossRef]
  29. Barbanti, M.; Yang, T.-H.; Rodès Cabau, J.; Tamburino, C.; Wood, D.A.; Jilaihawi, H.; Blanke, P.; Makkar, R.R.; Latib, A.; Colombo, A.; et al. Anatomical and Procedural Features Associated With Aortic Root Rupture During Balloon-Expandable Transcatheter Aortic Valve Replacement. Circulation 2013, 128, 244–253. [Google Scholar] [CrossRef] [PubMed]
  30. Girdauskas, E.; Owais, T.; Fey, B.; Kuntze, F.; Lauer, B.; Borger, M.A.; Conradi, L.; Reichenspurner, H.; Kuntze, T. Subannular perforation of left ventricular outflow tract associated with transcatheter valve implantation: Pathophysiological background and clinical implications. Eur. J. Cardiothorac. Surg. 2017, 51, 91–96. [Google Scholar] [CrossRef]
  31. Wang, Q.; Kodali, S.; Primiano, C.; Sun, W. Simulations of transcatheter aortic valve implantation: Implications for aortic root rupture. Biomech. Model. Mechanobiol. 2015, 14, 29–38. [Google Scholar] [CrossRef] [PubMed]
  32. Sá, M.P.; Jacquemyn, X.; Van den Eynde, J.; Tasoudis, P.; Erten, O.; Sicouri, S.; Macedo, F.Y.; Pasala, T.; Kaple, R.; Weymann, A.; et al. Impact of Paravalvular Leak on Outcomes After Transcatheter Aortic Valve Implantation: Meta-Analysis of Kaplan-Meier-derived Individual Patient Data. Struct. Heart 2023, 7, 100118. [Google Scholar] [CrossRef]
  33. Bianchi, M.; Marom, G.; Ghosh, R.P.; Rotman, O.M.; Parikh, P.; Gruberg, L.; Bluestein, D. Patient-specific simulation of transcatheter aortic valve replacement: Impact of deployment options on paravalvular leakage. Biomech. Model. Mechanobiol. 2019, 18, 435–451. [Google Scholar] [CrossRef]
  34. Luraghi, G.; Matas, J.F.R.; Beretta, M.; Chiozzi, N.; Iannetti, L.; Migliavacca, F. The impact of calcification patterns in transcatheter aortic valve performance: A fluid-structure interaction analysis. Comput. Methods Biomech. Biomed. Eng. 2021, 24, 375–383. [Google Scholar] [CrossRef]
  35. Luraghi, G.; Migliavacca, F.; Garcia-Gonzalez, A.; Chiastra, C.; Rossi, A.; Cao, D.; Stefanini, G.; Matas, J.F.R. On the Modeling of Patient-Specific Transcatheter Aortic Valve Replacement: A Fluid-Structure Interaction Approach. Cardiovasc. Eng. Technol. 2019, 10, 437–455. [Google Scholar] [CrossRef]
  36. Mao, W.; Wang, Q.; Kodali, S.; Sun, W. Numerical Parametric Study of Paravalvular Leak Following a Transcatheter Aortic Valve Deployment Into a Patient-Specific Aortic Root. J. Biomech. Eng. 2018, 140, 1010071–10100711. [Google Scholar] [CrossRef]
  37. Nappi, F.; Mazzocchi, L.; Spadaccio, C.; Attias, D.; Timofeva, I.; Macron, L.; Iervolino, A.; Morganti, S.; Auricchio, F. CoreValve vs. Sapien 3 Transcatheter Aortic Valve Replacement: A Finite Element Analysis Study. Bioengineering 2021, 8, 52. [Google Scholar] [CrossRef]
  38. Ghosh, R.P.; Marom, G.; Bianchi, M.; D’Souza, K.; Zietak, W.; Bluestein, D. Numerical evaluation of transcatheter aortic valve performance during heart beating and its post-deployment fluid-structure interaction analysis. Biomech. Model. Mechanobiol. 2020, 19, 1725–1740. [Google Scholar] [CrossRef]
  39. Ribeiro, H.B.; Webb, J.G.; Makkar, R.R.; Cohen, M.G.; Kapadia, S.R.; Kodali, S.; Tamburino, C.; Barbanti, M.; Chakravarty, T.; Jilaihawi, H.; et al. Predictive factors, management, and clinical outcomes of coronary obstruction following transcatheter aortic valve implantation: Insights from a large multicenter registry. J. Am. Coll. Cardiol. 2013, 62, 1552–1562. [Google Scholar] [CrossRef] [PubMed]
  40. Khan Jaffar, M.; Kamioka, N.; Lisko John, C.; Perdoncin, E.; Zhang, C.; Maini, A.; Chen, M.; Li, Y.; Ludwig, S.; Westermann, D.; et al. Coronary Obstruction From TAVR in Native Aortic Stenosis. JACC Cardiovasc. Interv. 2023, 16, 415–425. [Google Scholar] [CrossRef] [PubMed]
  41. Heitkemper, M.; Hatoum, H.; Azimian, A.; Yeats, B.; Dollery, J.; Whitson, B.; Rushing, G.; Crestanello, J.; Lilly, S.M.; Dasi, L.P. Modeling risk of coronary obstruction during transcatheter aortic valve replacement. J. Thorac. Cardiovasc. Surg. 2020, 159, 829–838.e3. [Google Scholar] [CrossRef]
  42. Holst, K.; Becker, T.; Magruder, J.T.; Yadav, P.; Stewart, J.; Rajagopal, V.; Liu, S.; Polsani, V.; Dasi, L.P.; Thourani, V.H. Beyond Static Planning: Computational Predictive Modeling to Avoid Coronary Artery Occlusion in TAVR. Ann. Thorac. Surg. 2025, 119, 145–151. [Google Scholar] [CrossRef]
  43. Wei, Z.A.; Sonntag, S.J.; Toma, M.; Singh-Gryzbon, S.; Sun, W. Computational Fluid Dynamics Assessment Associated with Transcatheter Heart Valve Prostheses: A Position Paper of the ISO Working Group. Cardiovasc. Eng. Technol. 2018, 9, 289–299. [Google Scholar] [CrossRef]
  44. Plitman Mayo, R.; Yaakobovich, H.; Finkelstein, A.; Shadden, S.C.; Marom, G. Numerical models for assessing the risk of leaflet thrombosis post-transcatheter aortic valve-in-valve implantation. R. Soc. Open Sci. 2020, 7, 201838. [Google Scholar] [CrossRef]
  45. Hatoum, H.; Maureira, P.; Lilly, S.; Dasi, L.P. Impact of BASILICA on Sinus and Neo-Sinus Hemodynamics after Valve-in-Valve with and without Coronary Flow. Cardiovasc. Revasc. Med. 2020, 21, 271–276. [Google Scholar] [CrossRef]
  46. Rocatello, G.; El Faquir, N.; De Santis, G.; Iannaccone, F.; Bosmans, J.; De Backer, O.; Sondergaard, L.; Segers, P.; De Beule, M.; de Jaegere, P.; et al. Patient-Specific Computer Simulation to Elucidate the Role of Contact Pressure in the Development of New Conduction Abnormalities After Catheter-Based Implantation of a Self-Expanding Aortic Valve. Circ. Cardiovasc. Interv. 2018, 11, e005344. [Google Scholar] [CrossRef]
  47. Wang, M.; Wang, Y.; Debusschere, N.; Rocatello, G.; Cheng, S.; Jin, J.; Yu, S. Predicting new-onset persistent conduction disturbance following transcatheter aortic valve replacement: The usefulness of FEOPS finite element analysis. BMC Cardiovasc. Disord. 2024, 24, 607. [Google Scholar] [CrossRef]
  48. Dowling, C.; Gooley, R.; McCormick, L.; Firoozi, S.; Brecker, S.J. Patient-specific Computer Simulation: An Emerging Technology for Guiding the Transcatheter Treatment of Patients with Bicuspid Aortic Valve. Interv. Cardiol. 2021, 16, e26. [Google Scholar] [CrossRef]
  49. Pasta, S.; Rinaudo, A.; Luca, A.; Pilato, M.; Scardulla, C.; Gleason, T.G.; Vorp, D.A. Difference in hemodynamic and wall stress of ascending thoracic aortic aneurysms with bicuspid and tricuspid aortic valve. J. Biomech. 2013, 46, 1729–1738. [Google Scholar] [CrossRef]
  50. de Jaegere, P.; Rocatello, G.; Prendergast, B.D.; de Backer, O.; Van Mieghem, N.M.; Rajani, R. Patient-specific computer simulation for transcatheter cardiac interventions: What a clinician needs to know. Heart 2019, 105 (Suppl. 2), s21–s27. [Google Scholar] [CrossRef]
  51. Kohli, K.; Wei, Z.A.; Sadri, V.; Easley, T.F.; Pierce, E.L.; Zhang, Y.N.; Wang, D.D.; Greenbaum, A.B.; Lisko, J.C.; Khan, J.M.; et al. Framework for Planning TMVR using 3-D Imaging, In Silico Modeling, and Virtual Reality. Struct. Heart 2020, 4, 336–341. [Google Scholar] [CrossRef]
  52. Lau, K.D.; Diaz, V.; Scambler, P.; Burriesci, G. Mitral valve dynamics in structural and fluid-structure interaction models. Med. Eng. Phys. 2010, 32, 1057–1064. [Google Scholar] [CrossRef]
  53. Dabiri, Y.; Yao, J.; Mahadevan, V.S.; Gruber, D.; Arnaout, R.; Gentzsch, W.; Guccione, J.M.; Kassab, G.S. Mitral Valve Atlas for Artificial Intelligence Predictions of MitraClip Intervention Outcomes. Front. Cardiovasc. Med. 2021, 8, 759675. [Google Scholar] [CrossRef]
  54. Dabiri, Y.; Mahadevan, V.S.; Guccione, J.M.; Kassab, G.S. Machine learning used for simulation of MitraClip intervention: A proof-of-concept study. Front. Genet. 2023, 14, 1142446. [Google Scholar] [CrossRef]
  55. Prescott, B.; Abunassar, C.J.; Baxevanakis, K.P.; Zhao, L. Computational evaluation of mitral valve repair with MitraClip. Vessel. Plus 2019, 3, 13. [Google Scholar] [CrossRef]
  56. Hart, E.A.; De Bock, S.; De Beule, M.; Teske, A.J.; Chamuleau, S.A.; Voskuil, M.; Kraaijeveld, A.O. Transoesophageal echocardiography-based computational simulation of the mitral valve for MitraClip placement. EuroIntervention 2019, 15, e239–e241. [Google Scholar] [CrossRef]
  57. Messika-Zeitoun, D.; Mousavi, J.; Pourmoazen, M.; Cotte, F.; Dreyfus, J.; Nejjari, M.; Attias, D.; Kloeckner, M.; Ghostine, S.; Pierrard, R.; et al. Computational simulation model of transcatheter edge-to-edge mitral valve repair: A proof-of-concept study. Eur. Heart J.-Cardiovasc. Imaging 2024, 25, 1415–1422. [Google Scholar] [CrossRef]
  58. Rogers Jason, H.; Low Reginald, I. Noncentral Mitral Regurgitation. JACC 2013, 62, 2378–2381. [Google Scholar] [CrossRef] [PubMed][Green Version]
  59. Wei, P.; Feng, S.; Zhang, F.; Li, H.; Zhuang, D.; Jiang, H.; Zhao, G.; Dong, J.; Wang, C.; Ouyang, W.; et al. Comparative Analysis of Central and Noncentral Degenerative Mitral Regurgitation Treated With Transcatheter Mitral Valve Edge-To-Edge Repair. Catheter. Cardiovasc. Interv. 2025, 105, 707–719. [Google Scholar] [CrossRef]
  60. Toma, M.; Bloodworth, C.H.; Pierce, E.L.; Einstein, D.R.; Cochran, R.P.; Yoganathan, A.P.; Kunzelman, K.S. Fluid-Structure Interaction Analysis of Ruptured Mitral Chordae Tendineae. Ann. Biomed. Eng. 2017, 45, 619–631. [Google Scholar] [CrossRef]
  61. Alkhouli, M.; Du, C.; Killu, A.; Simard, T.; Noseworthy, P.A.; Friedman, P.A.; Curtis, J.P.; Freeman, J.V.; Holmes, D.R. Clinical impact of residual leaks following left atrial appendage occlusion: Insights from the NCDR LAAO registry. JACC Clin. Electrophysiol. 2022, 8, 766–778. [Google Scholar] [CrossRef]
  62. Michiels, K.; Heffinck, E.; Astudillo, P.; Wong, I.; Mortier, P.; Bavo, A.M. Automated MSCT analysis for planning left atrial appendage occlusion using artificial intelligence. J. Interv. Cardiol. 2022, 2022, 5797431. [Google Scholar] [CrossRef]
  63. Bavo, A.M.; Wilkins, B.T.; Garot, P.; De Bock, S.; Saw, J.; Søndergaard, L.; De Backer, O.; Iannaccone, F. Validation of a computational model aiming to optimize preprocedural planning in percutaneous left atrial appendage closure. J. Cardiovasc. Comput. Tomogr. 2020, 14, 149–154. [Google Scholar] [CrossRef]
  64. Buysschaert, I.; Viaene, D. Clinical impact of preprocedural CT-based 3D computational simulation of left atrial appendage occlusion with amulet. J. Interv. Cardiol. 2021, 2021, 9972228. [Google Scholar] [CrossRef]
  65. Garot, P.; Iriart, X.; Aminian, A.; Kefer, J.; Freixa, X.; Cruz-Gonzalez, I.; Berti, S.; Rosseel, L.; Ibrahim, R.; Korsholm, K.; et al. Value of FEops HEARTguide patient-specific computational simulations in the planning of left atrial appendage closure with the Amplatzer Amulet closure device: Rationale and design of the PREDICT-LAA study. Open Heart 2020, 7, e001326. [Google Scholar] [CrossRef]
  66. Ranard, L.S.; Vahl, T.P.; Sommer, R.; Ng, V.; Leb, J.; Lehenbauer, K.; Sitticharoenchai, P.; Khalique, O.; Hamid, N.; De Beule, M.; et al. FEops HEARTguide Patient-Specific Computational Simulations for WATCHMAN FLX Left Atrial Appendage Closure: A Retrospective Study. JACC Adv. 2022, 1, 100139. [Google Scholar] [CrossRef]
  67. Medina, E.; Aguado, A.; Mill, J.; Freixa, X.; Arzamendi, D.; Yagüe, C.; Camara, O. VRIDAA: Virtual Reality Platform for Training and Planning Implantations of Occluder Devices in Left Atrial Appendages. VCBM 2020. [Google Scholar] [CrossRef]
  68. Masci, A.; Barone, L.; Dedè, L.; Fedele, M.; Tomasi, C.; Quarteroni, A.; Corsi, C. The impact of left atrium appendage morphology on stroke risk assessment in atrial fibrillation: A computational fluid dynamics study. Front. Physiol. 2019, 9, 1938. [Google Scholar] [CrossRef] [PubMed]
  69. Bosi, G.M.; Cook, A.; Rai, R.; Menezes, L.J.; Schievano, S.; Torii, R.; Burriesci, G. Computational fluid dynamic analysis of the left atrial appendage to predict thrombosis risk. Front. Cardiovasc. Med. 2018, 5, 34. [Google Scholar] [CrossRef] [PubMed]
  70. Otani, T.; Al-Issa, A.; Pourmorteza, A.; McVeigh, E.R.; Wada, S.; Ashikaga, H. A computational framework for personalized blood flow analysis in the human left atrium. Ann. Biomed. Eng. 2016, 44, 3284–3294. [Google Scholar] [CrossRef]
  71. Olivares, A.L.; Silva, E.; Nuñez-Garcia, M.; Butakoff, C.; Sánchez-Quintana, D.; Freixa, X.; Noailly, J.; de Potter, T.; Camara, O. In silico analysis of haemodynamics in patient-specific left atria with different appendage morphologies. In International Conference on Functional Imaging and Modeling of the Heart; Springer: Cham, Switzerland, 2017. [Google Scholar]
  72. Di Biase, L.; Santangeli, P.; Anselmino, M.; Mohanty, P.; Salvetti, I.; Gili, S.; Horton, R.; Sanchez, J.E.; Bai, R.; Mohanty, S.; et al. Does the left atrial appendage morphology correlate with the risk of stroke in patients with atrial fibrillation? Results from a multicenter study. J. Am. Coll. Cardiol. 2012, 60, 531–538. [Google Scholar] [CrossRef] [PubMed]
  73. Mill, J.; Agudelo, V.; Li, C.H.; Noailly, J.; Freixa, X.; Camara, O.; Arzamendi, D. Patient-specific flow simulation analysis to predict device-related thrombosis in left atrial appendage occluders. REC Interv. Cardiol. 2021, 3, 278–285. [Google Scholar] [CrossRef]
  74. Aguado, A.M.; Olivares, A.L.; Yagüe, C.; Silva, E.; Nuñez-García, M.; Fernandez-Quilez, Á.; Mill, J.; Genua, I.; Arzamendi, D.; De Potter, T.; et al. In silico optimization of left atrial appendage occluder implantation using interactive and modeling tools. Front. Physiol. 2019, 10, 237. [Google Scholar] [CrossRef] [PubMed]
  75. Vogl, B.; El Shaer, A.; Ponce, A.C.; Bavo, A.; De Beule, M.; Alkhouli, M.; Hatoum, H. TCT-380 Predicting Device-Related Thrombosis Using Computational Fluid Dynamics. J. Am. Coll. Cardiol. 2022, 80 (Suppl. 12), B153. [Google Scholar] [CrossRef]
  76. D’Alessandro, N.; Falanga, M.; Masci, A.; Severi, S.; Corsi, C. Preliminary findings on left atrial appendage occlusion simulations applying different endocardial devices. Front. Cardiovasc. Med. 2023, 10, 1067964. [Google Scholar] [CrossRef]
  77. Planas, E.; Mill, J.; Olivares, A.L.; Morales, X.; Pons, M.I.; Iriart, X.; Cochet, H.; Camara, O. In-silico analysis of device-related thrombosis for different left atrial appendage occluder settings. In International Workshop on Statistical Atlases and Computational Models of the Heart; Springer: Cham, Switzerland, 2022. [Google Scholar]
  78. Mendez, K.; Kennedy, D.; Wang, D.D.; O’nEill, B.; Roche, E.T. Left Atrial Appendage Occlusion: Current Stroke Prevention Strategies and a Shift Toward Data-Driven, Patient-Specific Approaches. J. Soc. Cardiovasc. Angiogr. Interv. 2022, 1, 100405. [Google Scholar] [CrossRef]
  79. Chatzizisis, Y.S.; Edelman, E.R. Revolutionizing Cardiovascular Interventions with Artificial Intelligence. J. Soc. Cardiovasc. Angiogr. Interv. 2025, 4, 102580. [Google Scholar] [CrossRef]
  80. Sun, S.; Yeh, L.; Imanzadeh, A.; Kooraki, S.; Kheradvar, A.; Bedayat, A. The Current Landscape of Artificial Intelligence in Imaging for Transcatheter Aortic Valve Replacement. Curr. Radiol. Rep. 2024, 12, 113–120. [Google Scholar] [CrossRef]
  81. Muralidharan, V.; Adewale, B.A.; Huang, C.J.; Nta, M.T.; Ademiju, P.O.; Pathmarajah, P.; Hang, M.K.; Adesanya, O.; Abdullateef, R.O.; Babatunde, A.O.; et al. A scoping review of reporting gaps in FDA-approved AI medical devices. NPJ Digit. Med. 2024, 7, 273. [Google Scholar] [CrossRef]
  82. Marey, A.; Arjmand, P.; Alerab, A.; Eslami, M.J.; Saad, A.M.; Sanchez, N.; Umair, M. Explainability, transparency and black box challenges of AI in radiology: Impact on patient care in cardiovascular radiology. Egypt. J. Radiol. Nucl. Med. 2024, 55, 183. [Google Scholar] [CrossRef]
  83. Alzubaidi, L.; Zhang, J.; Humaidi, A.J.; Al-Dujaili, A.; Duan, Y.; Al-Shamma, O.; Santamaría, J.; Fadhel, M.A.; Al-Amidie, M.; Farhan, L. Review of deep learning: Concepts, CNN architectures, challenges, applications, future directions. J. Big Data 2021, 8, 53. [Google Scholar] [CrossRef]
  84. van Zyl, C.; Ye, X.; Naidoo, R. Harnessing eXplainable artificial intelligence for feature selection in time series energy forecasting: A comparative analysis of Grad-CAM and SHAP. Appl. Energy 2024, 353, 122079. [Google Scholar] [CrossRef]
  85. Maloney, S.; Haines, T. Issues of cost-benefit and cost-effectiveness for simulation in health professions education. Adv. Simul. 2016, 1, 13. [Google Scholar] [CrossRef] [PubMed]
  86. Lee, K.S.; Klein, A.J.; Bricker, R.S.; Salavitabar, A.; Ijioma, N.N.; Indik, J.H.; Jayasuriya, S.N.; Fortuin, F.D.; Damluji, A.A.; Brilakis, E.S.; et al. Current State of Simulation in Interventional Cardiology Training: Results of a SCAI Survey. J. Soc. Cardiovasc. Angiogr. Interv. 2025, 4 Pt A, 102566. [Google Scholar] [CrossRef] [PubMed]
  87. Tolu-Akinnawo, O.Z.; Ezekwueme, F.; Omolayo, O.; Batheja, S.; Awoyemi, T. Advancements in Artificial Intelligence in Noninvasive Cardiac Imaging: A Comprehensive Review. Clin. Cardiol. 2025, 48, e70087. [Google Scholar] [CrossRef]
  88. Biondi-Zoccai, G.; D’Ascenzo, F.; Giordano, S.; Mirzoyev, U.; Erol, Ç.; Cenciarelli, S.; Leone, P.; Versaci, F. Artificial Intelligence in Cardiology: General Perspectives and Focus on Interventional Cardiology. Anatol. J. Cardiol. 2025, 29, 152–163. [Google Scholar] [CrossRef]
  89. Mastrodicasa, D.; van Assen, M.; Huisman, M.; Leiner, T.; Williamson, E.E.; Nicol, E.D.; Allen, B.D.; Saba, L.; Vliegenthart, R.; Hanneman, K. Use of AI in Cardiac CT and MRI: A Scientific Statement from the ESCR, EuSoMII, NASCI, SCCT, SCMR, SIIM, and RSNA. Radiology 2025, 314, e240516. [Google Scholar] [CrossRef]
  90. Zhong, L.; Zhang, J.-M.; Su, B.; Tan, R.S.; Allen, J.C.; Kassab, G.S. Application of Patient-Specific Computational Fluid Dynamics in Coronary and Intra-Cardiac Flow Simulations: Challenges and Opportunities. Front. Physiol. 2018, 9, 742. [Google Scholar] [CrossRef]
  91. Morris, P.D.; Narracott, A.; von Tengg-Kobligk, H.; Soto, D.A.S.; Hsiao, S.; Lungu, A.; Evans, P.; Bressloff, N.W.; Lawford, P.V.; Hose, D.R.; et al. Computational fluid dynamics modelling in cardiovascular medicine. Heart 2016, 102, 18–28. [Google Scholar] [CrossRef] [PubMed]
  92. Abdulkareem, M.; Brahier, M.S.; Zou, F.; Taylor, A.; Thomaides, A.; Bergquist, P.J.; Srichai, M.B.; Lee, A.M.; Vargas, J.D.; Petersen, S.E. Generalizable Framework for Atrial Volume Estimation for Cardiac CT Images Using Deep Learning with Quality Control Assessment. Front. Cardiovasc. Med. 2022, 9, 822269. [Google Scholar] [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Download PDF
Non-Contact Laser Therapy for Glaucoma: A Review of Direct Selective Laser Trabeculoplasty
Previous
Beyond Cultures: The Evolving Role of Molecular Diagnostics, Synovial Biomarkers and Artificial Intelligence in the Diagnosis of Prosthetic Joint Infections
Next