Computational Models in Cardiovascular System

Objective: Pericardial adhesions can unexpectedly occur prior to cardiac surgery or catheter ablation, even in patients without known risk factors, potentially increasing procedural risks. This study proposed and validated a novel, quantitative, and noninvasive method for detecting pericardial adhesions using four-dimensional computed tomography (4D CT). Methods: We evaluated preoperative 4D CT datasets from 20 patients undergoing cardiac surgery with and without pericardial adhesions. Our novel approach integrates expert-guided pericardial segmentation, symmetric diffeomorphic registration, and motion disparity analysis. The method quantifies tissue motion differences by computing the displacement fields between the pericardium and epicardial adipose tissue (EAT), with a particular focus on the left anterior descending (LAD) region. Results: Statistical analysis revealed significant differences between adhesion and non-adhesion groups (p < 0.01) using two newly developed metrics: peak ratio (PR) and distribution width index (DWI). Adhesion cases demonstrated characteristic high PR values (>100) with low DWI values (<0.3), while non-adhesion cases showed moderate PR values (<50) with higher DWI values (>0.4). Conclusions: This proof-of-concept study validated a novel quantitative framework for assessing pericardial adhesions using 4D CT imaging and provides an objective and computationally efficient tool for preoperative assessment in clinical settings. These findings suggest the potential clinical utility of this framework in surgical planning and risk assessment. Full article

This study proposes an innovative mathematical model for assessing microcirculation in patients with diabetic ulcers, using the ankle–brachial index (ABI). The methodology combines Bond Graph (BG) modeling and Particle Swarm Optimization (PSO), enabling a detailed analysis of hemodynamic patterns in a pilot sample of three patients. The results revealed a correlation between ulcer size and reduced ABI values, suggesting that deficits in microcirculation directly impact the severity of lesions. Furthermore, despite variations in ABI values and arterial pressures, all patients exhibited high capillary resistance, indicating difficulties in microcirculatory blood flow. The PSO-optimized parameters for the capillary equivalent circuit were found to be $R_1=89.784\Omega$, $R_2=426.55\Omega$, $L=27.506\mathrm{H}$, and $C=0.00040675\mathrm{F}$, which confirms the presence of high vascular resistance and reduced compliance in the microvascular system of patients with diabetic foot ulcers. This quantitative analysis, made possible through mathematical modeling, is crucial for detecting subtle changes in microcirculatory dynamics, which may not be easily identified through conventional pressure measurements alone. The increased capillary resistance observed may serve as a key indicator of vascular impairment, potentially guiding early intervention strategies and optimizing diabetic ulcer treatment. We acknowledge that the sample size of three patients represents a limitation of the study, but this number was intentionally chosen to allow for a detailed and controlled analysis of the variables involved. Although the findings are promising, additional experimental validations are necessary to confirm the clinical applicability of the model in a larger patient sample, thus solidifying its relevance in clinical practice. Full article

Objectives: Early detection of cardiovascular diseases and their pre-existing conditions, arteriosclerosis and atherosclerosis, is crucial to increasing a patient’s chance of survival. While imaging technologies and invasive procedures provide a reliable diagnosis, they carry high costs and risks for patients. This study aims to explore impedance plethysmography (IPG) as a non-invasive and affordable alternative for diagnosis. Methods: To address the current lack of large-scale, high-quality impedance data, we introduce arterioscope.sim, a simulation platform that models arterial blood flow and computes the electrical conductivity of blood. The platform simulates bioimpedance measurements on specific body segments using patient-specific parameters. The study investigates how introducing arterial diseases into the simulation affects the bioimpedance signals. Results: The simulation results demonstrate that introducing atherosclerosis and arteriosclerosis leads to significant changes in the computed signals compared to simulations of healthy arteries. Furthermore, simulation of a patient-specific healthy artery strongly correlates with measured signals from a healthy volunteer. Conclusions and significance: arterioscope.sim effectively simulates bioimpedance signals in healthy and diseased arteries and highlights the potential of using these signals for early diagnosis of arterial diseases, offering a non-invasive and cost-effective alternative to traditional diagnostic methods. Full article

Silent Brain Infarction (SBI) is increasingly recognized in patients with cardiac conditions, particularly Atrial Fibrillation (AF) in elderly patients and those undergoing Transcatheter Aortic Valve Implantation (TAVI). While these infarcts often go unnoticed due to a lack of acute symptoms, they are associated with a threefold increase in stroke risk and are considered a precursor to ischemic stroke. Moreover, accumulating evidence suggests that SBI may contribute to the development of dementia, depression, and cognitive decline, particularly in the elderly population. The burden of SBI is substantial, with studies showing that up to 11 million Americans may experience a silent stroke annually. In AF patients, silent brain infarcts are common and can lead to progressive brain damage, even in those receiving anticoagulation therapy. The use of cerebral embolic protection devices (CEPDs) during TAVI has been explored to mitigate the risk of stroke; however, their efficacy remains under debate. Despite advancements in TAVI technology, cerebrovascular events, including silent brain lesions, continue to pose significant challenges, underscoring the need for improved preventive strategies and therapeutic approaches. We propose a device consisting of a strut structure placed at the base of the treated artery to model the potential risk of cerebral embolisms caused by atrial fibrillation, thromboembolism, or dislodged debris of varying potential TAVI patients. The study has been carried out in two stages. Both are based on computational fluid dynamics (CFD) coupled with the Lagrangian particle tracking method. The first stage of the work evaluates a variety of strut thicknesses and inter-strut spacings, contrasting with the device-free baseline geometry. The analysis is carried out by imposing flow rate waveforms characteristic of healthy and AF patients. Boundary conditions are calibrated to reproduce physiological flow rates and pressures in a patient’s aortic arch. In the second stage, the optimal geometric design from the first stage was employed, with the addition of lateral struts to prevent the filtration of particles and electronegatively charged strut surfaces, studying the effect of electrical forces on the clots if they are considered charged. Flowrate boundary conditions were used to emulate both healthy and AF conditions. Results from numerical simulations coming from the first stage indicate that the device blocks particles of sizes larger than the inter-strut spacing. It was found that lateral strut space had the highest impact on efficacy. Based on the results of the second stage, deploying the electronegatively charged device in all three aortic arch arteries, the number of particles entering these arteries was reduced on average by $62.6$% and $51.2$%, for the healthy and diseased models respectively, matching or surpassing current oral anticoagulant efficacy. In conclusion, the device demonstrated a two-fold mechanism for filtering emboli: (1) while the smallest particles are deflected by electrostatic repulsion, avoiding micro embolisms, which could lead to cognitive impairment, the largest ones are mechanically filtered since they cannot fit in between the struts, effectively blocking the full range of particle sizes analyzed in this study. The device presented in this manuscript offers an anticoagulant-free method to prevent stroke and SBIs, imperative given the growing population of AF and elderly patients. Full article

Analyzing heart sound signals presents a novel approach for early diagnosis of pediatric congenital heart disease. The existing segmentation algorithms have limitations in accurately distinguishing the first (S1) and second (S2) heart sounds, limiting the diagnostic utility of cardiac cycle data for pediatric pathology assessment. This study proposes a time bidirectional long short-term memory network (TBLSTM) based on multi-scale analysis to segment pediatric heart sound signals according to different cardiac cycles. Mel frequency cepstral coefficients and dynamic characteristics of the heart sound fragments were extracted and input into random forest for multi-classification of congenital heart disease. The segmentation model achieved an overall F1 score of 94.15% on the verification set, with specific F1 scores of 90.25% for S1 and 86.04% for S2. In a situation where the number of cardiac cycles in the heart sound fragments was set to six, the results for multi-classification achieved stabilization. The performance metrics for this configuration were as follows: accuracy of 94.43%, sensitivity of 95.58%, and an F1 score of 94.51%. Furthermore, the segmentation model demonstrates robustness in accurately segmenting pediatric heart sound signals across different heart rates and in the presence of noise. Notably, the number of cardiac cycles in heart sound fragments directly impacts the multi-classification of these heart sound signals. Full article

Three-dimensional echocardiography (3D ECHO) and magnetic resonance (MR) imaging are frequently used in patients and animals to evaluate heart functions. Inverse finite element (FE) modeling is increasingly applied to MR images to quantify left ventricular (LV) function and estimate myocardial contractility and other cardiac biomarkers. It remains unclear, however, as to whether myocardial contractility derived from the inverse FE model based on 3D ECHO images is comparable to that derived from MR images. To address this issue, we developed a subject-specific inverse FE model based on 3D ECHO and MR images acquired from seven healthy swine models to investigate if there are differences in myocardial contractility and LV geometrical features derived using these two imaging modalities. We showed that end-systolic and end-diastolic volumes derived from 3D ECHO images are comparable to those derived from MR images ($R^2=0.805$ and $0.969$, respectively). As a result, ejection fraction from 3D ECHO and MR images are linearly correlated ($R^2=0.977$) with the limit of agreement (LOA) ranging from −17.95% to 45.89%. Using an inverse FE modeling to fit pressure and volume waveforms in subject-specific LV geometry reconstructed from 3D ECHO and MR images, we found that myocardial contractility derived from these two imaging modalities are linearly correlated with an $R^2$ value of 0.989, a gradient of 0.895, and LOA ranging from −6.11% to 36.66%. This finding supports using 3D ECHO images in image-based inverse FE modeling to estimate myocardial contractility. Full article

During systole, longitudinal shortening of the left ventricle (LV) displaces the aortic root toward the apex of the heart and stretches the ascending aorta (AA). An in silico study (Living Left Heart Human Model, Dassault Systèmes Simulia Corporation) demonstrated that stiffening of the AA affects myocardial stress and LV strain patterns. With AA stiffening, myofiber stress increased overall in the LV, with particularly high-stress areas at the septum. The most pronounced reduction in strain was noted along the septal longitudinal region. The pressure–volume loops showed that AA stiffening caused a deterioration in LV function, with increased end-systolic volume, reduced systolic LV pressure, decreased stroke volume and effective stroke work, but elevated end-diastolic pressure. An increase in myofiber contractility indicated that stroke volume and effective stroke work could be recovered, with an increase in LV end-systolic pressure and a decrease in end-diastolic pressure. Longitudinal and radial strains remained reduced, but circumferential strains increased over baseline, compensating for lost longitudinal LV function. Myofiber stress increased overall, with the most dramatic increase in the septal region and the LV apex. We demonstrate a direct mechanical pathophysiologic link between stiff AA and reduced longitudinal left ventricular strain which are common in patients with HFpEF. Full article

Research on abdominal aortic aneurysms (AAAs) primarily focuses on developing a clear understanding of the initiation, progression, and treatment of AAA through improved model accuracy. High-fidelity hemodynamic and biomechanical predictions are essential for clinicians to optimize preoperative planning and minimize therapeutic risks. Computational fluid dynamics (CFDs), finite element analysis (FEA), and fluid-structure interaction (FSI) are widely used to simulate AAA hemodynamics and biomechanics. However, the accuracy of these simulations depends on the utilization of realistic and sophisticated boundary conditions (BCs), which are essential for properly integrating the AAA with the rest of the cardiovascular system. Recent advances in machine learning (ML) techniques have introduced faster, data-driven surrogates for AAA modeling. These approaches can accelerate segmentation, predict hemodynamics and biomechanics, and assess disease progression. However, their reliability depends on high-quality training data derived from CFDs and FEA simulations, where BC modeling plays a crucial role. Accurate BCs can enhance ML predictions, increasing the clinical applicability. This paper reviews existing BC models, discussing their limitations and technical challenges. Additionally, recent advancements in ML and data-driven techniques are explored, discussing their current states, future directions, common algorithms, and limitations. Full article