Numerical modeling of hemodynamics scenarios of patient-specific coronary artery bypass grafts

Solving the Inverse Problem of Electrocardiography for Cardiac Digital Twins: A Survey

Cardiac digital twins (CDTs) are personalized virtual representations used to understand complex cardiac mechanisms. A critical component of CDT development is solving the ECG inverse problem, which enables the reconstruction of cardiac sources and the estimation of patient-specific electrophysiology (EP) parameters from surface ECG data. Despite challenges from complex cardiac anatomy, noisy ECG data, and the ill-posed nature of the inverse problem, recent advances in computational methods have greatly improved the accuracy and efficiency of ECG inverse inference, strengthening the fidelity of CDTs. This paper aims to provide a comprehensive review of the methods for solving ECG inverse problems, their validation strategies, their clinical applications, and their future perspectives. For the methodologies, we broadly classify state-of-the-art approaches into two categories: deterministic and probabilistic methods, including both conventional and deep learning-based techniques. Integrating physics laws with deep learning models holds promise, but challenges such as capturing dynamic electrophysiology accurately, accessing accurate domain knowledge, and quantifying prediction uncertainty persist. Integrating models into clinical workflows while ensuring interpretability and usability for healthcare professionals is essential. Overcoming these challenges will drive further research in CDTs.

Cardiovascular care with digital twin technology in the era of generative artificial intelligence

Digital twins, which are in silico replications of an individual and its environment, have advanced clinical decision-making and prognostication in cardiovascular medicine. The technology enables personalized simulations of clinical scenarios, prediction of disease risk, and strategies for clinical trial augmentation. Current applications of cardiovascular digital twins have integrated multi-modal data into mechanistic and statistical models to build physiologically accurate cardiac replicas to enhance disease phenotyping, enrich diagnostic workflows, and optimize procedural planning. Digital twin technology is rapidly evolving in the setting of newly available data modalities and advances in generative artificial intelligence, enabling dynamic and comprehensive simulations unique to an individual. These twins fuse physiologic, environmental, and healthcare data into machine learning and generative models to build real-time patient predictions that can model interactions with the clinical environment to accelerate personalized patient care. This review summarizes digital twins in cardiovascular medicine and their potential future applications by incorporating new personalized data modalities. It examines the technical advances in deep learning and generative artificial intelligence that broaden the scope and predictive power of digital twins. Finally, it highlights the individual and societal challenges as well as ethical considerations that are essential to realizing the future vision of incorporating cardiology digital twins into personalized cardiovascular care.

A reduced order model formulation for left atrium flow: an atrial fibrillation case

A data-driven reduced order model (ROM) based on a proper orthogonal decomposition-radial basis function (POD-RBF) approach is adopted in this paper for the analysis of blood flow dynamics in a patient-specific case of atrial fibrillation (AF). The full order model (FOM) is represented by incompressible Navier-Stokes equations, discretized with a finite volume (FV) approach. Both the Newtonian and the Casson's constitutive laws are employed. The aim is to build a computational tool able to efficiently and accurately reconstruct the patterns of relevant hemodynamics indices related to the stasis of the blood in a physical parametrization framework including the cardiac output in the Newtonian case and also the plasma viscosity and the hematocrit in the non-Newtonian one. Many FOM-ROM comparisons are shown to analyze the performance of our approach as regards errors and computational speed-up.

Accelerated simulation methodologies for computational vascular flow modelling

Vascular flow modelling can improve our understanding of vascular pathologies and aid in developing safe and effective medical devices. Vascular flow models typically involve solving the nonlinear Navier-Stokes equations in complex anatomies and using physiological boundary conditions, often presenting a multi-physics and multi-scale computational problem to be solved. This leads to highly complex and expensive models that require excessive computational time. This review explores accelerated simulation methodologies, specifically focusing on computational vascular flow modelling. We review reduced order modelling (ROM) techniques like zero-/one-dimensional and modal decomposition-based ROMs and machine learning (ML) methods including ML-augmented ROMs, ML-based ROMs and physics-informed ML models. We discuss the applicability of each method to vascular flow acceleration and the effectiveness of the method in addressing domain-specific challenges. When available, we provide statistics on accuracy and speed-up factors for various applications related to vascular flow simulation acceleration. Our findings indicate that each type of model has strengths and limitations depending on the context. To accelerate real-world vascular flow problems, we propose future research on developing multi-scale acceleration methods capable of handling the significant geometric variability inherent to such problems.

Reduced Order Modeling for Real-Time Stent Deformation Simulations of Transcatheter Aortic Valve Prostheses

Computational modeling can be a critical tool to predict deployment behavior for transcatheter aortic valve replacement (TAVR) in patients with aortic stenosis. However, due to the mechanical complexity of the aortic valve and the multiphysics nature of the problem, described by partial differential equations (PDEs), traditional finite element (FE) modeling of TAVR deployment is computationally expensive. In this preliminary study, a PDEs-based reduced order modeling (ROM) framework is introduced for rapidly simulating structural deformation of the Medtronic Evolut R valve stent frame. Using fifteen probing points from an Evolut model with parametrized loads enforced, 105 FE simulations were performed in the so-called offline phase, creating a snapshot library. The library was used in the online phase of the ROM for a new set of applied loads via the proper orthogonal decomposition-Galerkin (POD-Galerkin) approach. Simulations of small radial deformations of the Evolut stent frame were performed and compared to full order model (FOM) solutions. Linear elastic and hyperelastic constitutive models in steady and unsteady regimes were implemented within the ROM. Since the original POD-Galerkin method is formulated for linear problems, specific methods for the nonlinear terms in the hyperelastic case were employed, namely, the Discrete Empirical Interpolation Method. The ROM solutions were in strong agreement with the FOM in all numerical experiments, with a speed-up of at least 92% in CPU Time. This framework serves as a first step toward real-time predictive models for TAVR deployment simulations.

Artificial Intelligence, Computational Simulations, and Extended Reality in Cardiovascular Interventions

Artificial intelligence, computational simulations, and extended reality, among other 21st century computational technologies, are changing the health care system. To collectively highlight the most recent advances and benefits of artificial intelligence, computational simulations, and extended reality in cardiovascular therapies, we coined the abbreviation AISER. The review particularly focuses on the following applications of AISER: 1) preprocedural planning and clinical decision making; 2) virtual clinical trials, and cardiovascular device research, development, and regulatory approval; and 3) education and training of interventional health care professionals and medical technology innovators. We also discuss the obstacles and constraints associated with the application of AISER technologies, as well as the proposed solutions. Interventional health care professionals, computer scientists, biomedical engineers, experts in bioinformatics and visualization, the device industry, ethics committees, and regulatory agencies are expected to streamline the use of AISER technologies in cardiovascular interventions and medicine in general.

Personalized surgical planning for coronary bypass graft configurations using patient-specific computational modeling to avoid flow competition in arterial grafts

Objectives

Flow competition between coronary artery bypass grafts (CABG) and native coronary arteries is a significant problem affecting arterial graft patency. The objectives of this study were to compare the predictive hemodynamic flow resulting from various total arterial grafting configurations and to evaluate whether the use of computational fluid dynamics (CFD) models capable of predicting flow can assist surgeons to make better decisions for individual patients by avoiding poorly functioning grafts.

Methods

Sixteen cardiac surgeons declared their preferred CABG configuration using bilateral internal mammary and radial arteries for each of 5 patients who had differing degrees of severe triple vessel coronary disease. Surgeons selected both a preferred 'aortic' strategy, with at least one graft arising from the ascending aorta, and a preferred "anaortic" strategy which could be performed as a "no-aortic touch" operation. CT coronary angiograms of the 5 patients were coupled to CFD models using a novel flow solver "COMCAB." Twelve different CABG configurations were compared for each patient of which 4 were "aortic" and 8 were "anaortic." Surgeons then selected their preferred grafting configurations after being shown predictive hemodynamic metrics including functional assessment of stenoses (instantaneous wave-free ratio; fractional flow reserve), transit time flowmetry graft parameters (mean graft flow; pulsatility index) and myocardial perfusion.

Results

A total of 87.5% (7/8) of "anaortic" configurations compared to 25% (1/4) of "aortic" configurations led to unsatisfactory grafts in at least 1 of the 5 patients (P = 0.038). The use of the computational models led to a significant decrease in the selection of unsatisfactory grafting configurations when surgeons employed "anaortic" (21.25% (17/80) vs. 1.25% (1/80), P < 0.001) but not "aortic" techniques (5% (4/80) vs. 0% (0/80), P = 0.64). Similarly, there was an increase in the selection of ideal configurations for "anaortic" (6.25% (5/80) vs. 28.75% (23/80), P < 0.001) but not "aortic" techniques (65% (52/80) vs. 61.25% (49/80), P = 0.74). Furthermore, surgeons who planned to use more than one unique "anaortic" configuration across all 5 patients increased (12.5% (2/16) vs. 87.5% (14/16), P<0.001).

Conclusions

"COMCAB" is a promising tool to improve personalized surgical planning particularly for CABG configurations involving composite or sequential grafts which are used more frequently in anaortic operations.

Reduced order methods for parametric optimal flow control in coronary bypass grafts, toward patient-specific data assimilation

Coronary artery bypass grafts (CABG) surgery is an invasive procedure performed to circumvent partial or complete blood flow blockage in coronary artery disease. In this work, we apply a numerical optimal flow control model to patient-specific geometries of CABG, reconstructed from clinical images of real-life surgical cases, in parameterized settings. The aim of these applications is to match known physiological data with numerical hemodynamics corresponding to different scenarios, arisen by tuning some parameters. Such applications are an initial step toward matching patient-specific physiological data in patient-specific vascular geometries as best as possible. Two critical challenges that reportedly arise in such problems are: (a) lack of robust quantification of meaningful boundary conditions required to match known data as best as possible and (b) high computational cost. In this work, we utilize unknown control variables in the optimal flow control problems to take care of the first challenge. Moreover, to address the second challenge, we propose a time-efficient and reliable computational environment for such parameterized problems by projecting them onto a low-dimensional solution manifold through proper orthogonal decomposition-Galerkin.

Low Wall Shear Stress Is Associated with Saphenous Vein Graft Stenosis in Patients with Coronary Artery Bypass Grafting

Biomechanical forces may play a key role in saphenous vein graft (SVG) disease after coronary artery bypass graft (CABG) surgery. Computed tomography angiography (CTA) of 430 post-CABG patients were evaluated and 15 patients were identified with both stenosed and healthy SVGs for paired analysis. The stenosis was virtually removed, and detailed 3D models were reconstructed to perform patient-specific computational fluid dynamic (CFD) simulations. Models were processed to compute anatomic parameters, and hemodynamic parameters such as local and vessel-averaged wall shear stress (WSS), normalized WSS (WSS*), low shear area (LSA), oscillatory shear index (OSI), and flow rate. WSS* was significantly lower in pre-diseased SVG segments compared to corresponding control segments without disease (1.22 vs. 1.73, p = 0.012) and the area under the ROC curve was 0.71. No differences were observed in vessel-averaged anatomic or hemodynamic parameters between pre-stenosed and control whole SVGs. There are currently no clinically available tools to predict SVG failure post-CABG. CFD modeling has the potential to identify high-risk CABG patients who may benefit from more aggressive medical therapy and closer surveillance. Graphical Abstract.

Computational Evaluation of Venous Graft Geometries in Coronary Artery Bypass Surgery

Cardiothoracic surgeons are faced with a choice of different revascularization techniques and diameters for saphenous vein grafts (SVG) in coronary artery bypass graft surgery . Using computational simulations, we virtually investigate the effect of SVG geometry on hemodynamics of both venous grafts and the target coronary arteries. We generated patient-specific 3-dimensional anatomic models of coronary artery bypass graft surgery patients and quantified mechanical stimuli. We performed virtual surgery on 3 patient-specific models by modifying the geometry vein grafts to reflect single, Y, and sequential surgical configurations with SVG diameters ranging from 2 mm to 5 mm. Our study demonstrates that the coronary artery runoffs are relatively insensitive to the choice of SVG revascularization geometry. We observe a 10% increase in runoff when the SVG diameter is changed from 2 mm to 5 mm. The wall shear stress of SVG increases dramatically when the diameter drops, following an inverse power scaling with diameter. For a fixed diameter, the average wall shear stress on the vein graft varies in ascending order as single, Y, and sequential graft in the patient cohort. The runoff to the target coronary arteries changes marginally due to the choice of graft configuration or diameter. The shear stress on the vein graft depends on both flow rate and diameter and follows an inverse power scaling consistent with a Poiseuille flow assumption. Given the similarity in runoff with different surgical configurations, choices of SVG geometries can be informed by propensity for graft failure using shear stress evaluations.

Patient-Specific Hemodynamics of New Coronary Artery Bypass Configurations

PURPOSE

This study aims to quantify the patient-specific hemodynamics of complex conduit routing configurations of coronary artery bypass grafting (CABG) operation which are specifically suitable for off-pump surgeries. Coronary perfusion efficacy and local hemodynamics of multiple left internal mammary artery (LIMA) with sequential and end-to-side anastomosis are investigated. Using a full anatomical model comprised of aortic arch and coronary artery branches the optimum perfusion configuration in multi-vessel coronary artery stenosis is desired.

METHODOLOGY

Two clinically relevant CABG configurations are created using a virtual surgical planning tool where for each configuration set, the stenosis level, anastomosis distance and angle were varied. A non-Newtonian computational fluid dynamics solver in OpenFOAM incorporated with resistance boundary conditions representing the coronary perfusion physiology was developed. The numerical accuracy is verified and results agreed well with a validated commercial cardiovascular flow solver and experiments. For segmental performance analysis, new coronary perfusion indices to quantify deviation from the healthy scenario were introduced.

RESULTS

The first simulation configuration set;-a CABG targeting two stenos sites on the left anterior descending artery (LAD), the LIMA graft was capable of 31 mL/min blood supply for all the parametric cases and uphold the healthy LAD perfusion in agreement with the clinical experience. In the second end-to-side anastomosed graft configuration set;-the radial artery graft anastomosed to LIMA, a maximum of 64 mL/min flow rate in LIMA was observed. However, except LAD, the obtuse marginal (OM) and second marginal artery (m2) suffered poor perfusion. In the first set, average wall shear stress (WSS) were in the range of 4 to 35 dyns/cm² for in LAD. Nevertheless, for second configuration sets the WSS values were higher as the LIMA could not supply enough blood to OM and m2.

CONCLUSION

The virtual surgical configurations have the potential to improve the quality of operation by providing quantitative surgical insight. The degree of stenosis is a critical factor in terms of coronary perfusion and WSS. The sequential anastomosis can be done safely if the anastomosis angle is less than 90 degrees regardless of degree of stenosis. The smaller proposed perfusion index value, O(0.04 - 0) × 10², enable us to quantify the post-op hemodynamic performance by comparing with the ideal healthy physiological flow.

Using parametric model order reduction for inverse analysis of large nonlinear cardiac simulations

Predictive high-fidelity finite element simulations of human cardiac mechanics commonly require a large number of structural degrees of freedom. Additionally, these models are often coupled with lumped-parameter models of hemodynamics. High computational demands, however, slow down model calibration and therefore limit the use of cardiac simulations in clinical practice. As cardiac models rely on several patient-specific parameters, just one solution corresponding to one specific parameter set does not at all meet clinical demands. Moreover, while solving the nonlinear problem, 90% of the computation time is spent solving linear systems of equations. We propose to reduce the structural dimension of a monolithically coupled structure-Windkessel system by projection onto a lower-dimensional subspace. We obtain a good approximation of the displacement field as well as of key scalar cardiac outputs even with very few reduced degrees of freedom, while achieving considerable speedups. For subspace generation, we use proper orthogonal decomposition of displacement snapshots. Following a brief comparison of subspace interpolation methods, we demonstrate how projection-based model order reduction can be easily integrated into a gradient-based optimization. We demonstrate the performance of our method in a real-world multivariate inverse analysis scenario. Using the presented projection-based model order reduction approach can significantly speed up model personalization and could be used for many-query tasks in a clinical setting.

Longitudinal computational fluid dynamics study of stenosis and aneurysmal degeneration of an aortorenal bypass

Saphenous vein graft (SVG) bypass placement is regarded as the optimal option for renal artery stenosis, which usually causes secondary hypertension and poor renal perfusion. Using computational fluid dynamics, this study aimed to investigate the underlying hemodynamic mechanism of the vein aneurysm and stenosis after aortorenal bypass surgery. Three-dimensional models were reconstructed based on computed tomographic angiography images of a 20-year-old female patient who suffered from uncontrollable hypertension using the image processing package Mimics (Materialise). The morphology and hemodynamic parameters in the healthy state, at initial presentation and at post-operative 9-month and 2-year follow-ups after surgery were analysed. The hemodynamic parameters became normal in the left and right renal arteries after bypass surgery. However, flow separation and stagnation occurred at the post-operative 9-month aorta-vein anastomosis, which caused asymmetrical flow and extremely high wall shear stress (WSS) and WSS gradients at the outflow vein tract, where the stenosis occurred 2 years later. In addition, the graft bending produced an asymmetrical flow pattern downstream. This research revealed that the abnormal hemodynamics, including flow separation and extremely high WSS values and gradients, caused by the retrograde flow of aortorenal bypass may be responsible for the SVG degeneration. In addition, flow asymmetry due to vessel bending is a potential risk factor for SVG aneurysm dilation.

NUMERICAL INVESTIGATION OF THE EFFECTS OF BED SHAPE ON THE END-TO-SIDE CABG HEMODYNAMICS

Disrupted flow initiates and aggravates intimal thickening in the end-to-side (ETS) coronary artery bypass grafting (CABG), which may lead to failure. To enhance the post-intervention hemodynamics, the geometry is either optimized or totally reconfigured. Majority of configurations proposed by researchers have not suited CABG surgery, for they entailed rigorous manipulation on conventional grafts in situ, which was neither swift nor straightforward. The aim of the present study is, thus, to introduce a slight, yet effective, modification to a conventional ETS CABG configuration, and numerically investigate its effects on updated hemodynamic and structural environment, anticipating the longevity of proposed configuration and CABG success. This fairly simple modification may easily be made positioning a pre-designed anastomotic device between the bed of host artery in the conventional ETS CABG and its surrounding tissues. Conducting comprehensive numerical simulations, performance of the proposed configuration was assessed using idealized and patient-specific geometries of the conventional ETS CABG. Blood flow was simulated in a conventional and an updated CABG configuration considering 2-way fluid–structure interaction. Results revealed that, although the proposed configuration may induce higher structural stresses in vessels walls, it may improve important hemodynamic metrics such as wall shear stress gradient, oscillatory shear index, and relative residence time on host artery bed reducing disruption of flow. This study may also set the stage for design engineers and regulatory officials to evolve ETS CABG toward more hemodynamics-friendly approaches. Further in vitro, preclinical, and clinical experiments are, yet, entailed to accomplish ideal designs of procedural guidelines/grafts.