CRIMSON: An open-source software framework for cardiovascular integrated modelling and simulation

Arthurs, Christopher J.; Khlebnikov, Rostislav; Melville, Alex; Marčan, Marija; Gómez, Alberto; Dillon-Murphy, Desmond; Cuomo, Federica; Vieira, Miguel Silva; Schollenberger, Jonas; Lynch, Sabrina R.; Tossas-Betancourt, Christopher; Iyer, Kritika; Hopper, Sara; Livingston, Elizabeth; Youssefi, Pouya; Noorani, Alia; Ahmed, Sabrina Ben; Nauta, Foeke J.H.; Bakel, Theodorus M. J. van; Ahmed, Yunus; Bakel, Petrus Antonius Jacobus van; Mynard, Jonathan P.; Achille, Paolo Di; Gharahi, Hamidreza; Lau, K C; Filonova, Vasilina; Aguirre, Miquel; Nama, Nitesh; Xiao, Nan; Baek, Seungik; Garikipati, Krishna; Sahni, Onkar; Nordsletten, David; Figueroa, C. Alberto

doi:10.1371/journal.pcbi.1008881

Cited by 50 publications

(32 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The key computational modeling tasks, namely threedimensional anatomical reconstruction, mesh generation, boundary condition specification, and finite element analysis were performed using the validated open-source computational hemodynamics framework CRIMSON (Arthurs et al, 2021).…”

Section: Computational Modelingmentioning

confidence: 99%

A Combined Computational Fluid Dynamics and Arterial Spin Labeling MRI Modeling Strategy to Quantify Patient-Specific Cerebral Hemodynamics in Cerebrovascular Occlusive Disease

Schollenberger

Osborne

Hernandez‐García

et al. 2021

Front. Bioeng. Biotechnol.

Self Cite

View full text Add to dashboard Cite

Cerebral hemodynamics in the presence of cerebrovascular occlusive disease (CVOD) are influenced by the anatomy of the intracranial arteries, the degree of stenosis, the patency of collateral pathways, and the condition of the cerebral microvasculature. Accurate characterization of cerebral hemodynamics is a challenging problem. In this work, we present a strategy to quantify cerebral hemodynamics using computational fluid dynamics (CFD) in combination with arterial spin labeling MRI (ASL). First, we calibrated patient-specific CFD outflow boundary conditions using ASL-derived flow splits in the Circle of Willis. Following, we validated the calibrated CFD model by evaluating the fractional blood supply from the main neck arteries to the vascular territories using Lagrangian particle tracking and comparing the results against vessel-selective ASL (VS-ASL). Finally, the feasibility and capability of our proposed method were demonstrated in two patients with CVOD and a healthy control subject. We showed that the calibrated CFD model accurately reproduced the fractional blood supply to the vascular territories, as obtained from VS-ASL. The two patients revealed significant differences in pressure drop over the stenosis, collateral flow, and resistance of the distal vasculature, despite similar degrees of clinical stenosis severity. Our results demonstrated the advantages of a patient-specific CFD analysis for assessing the hemodynamic impact of stenosis.

show abstract

Section: Computational Modelingmentioning

confidence: 99%

A Combined Computational Fluid Dynamics and Arterial Spin Labeling MRI Modeling Strategy to Quantify Patient-Specific Cerebral Hemodynamics in Cerebrovascular Occlusive Disease

Schollenberger

Osborne

Hernandez‐García

et al. 2021

Front. Bioeng. Biotechnol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…FSI models were built with the open source computational hemodynamics platform CRIMSON ( Arthurs et al, 2021 ). Data from imaging, hemodynamics, and wall mechanics were combined to create FSI models comprised of 3D anatomical models having spatially varying anisotropic wall mechanical properties and external tissue support, with inlet flow waveforms and 3-element Windkessel models on each outflow branch.…”

Section: Methodsmentioning

confidence: 99%

Comparative Study of Human and Murine Aortic Biomechanics and Hemodynamics in Vascular Aging

et al. 2021

Self Cite

View full text Add to dashboard Cite

Introduction: Aging has many effects on the cardiovascular system, including changes in structure (aortic composition, and thus stiffening) and function (increased proximal blood pressure, and thus cardiac afterload). Mouse models are often used to gain insight into vascular aging and mechanisms of disease as they allow invasive assessments that are impractical in humans. Translation of results from murine models to humans can be limited, however, due to species-specific anatomical, biomechanical, and hemodynamic differences. In this study, we built fluid-solid-interaction (FSI) models of the aorta, informed by biomechanical and imaging data, to compare wall mechanics and hemodynamics in humans and mice at two equivalent ages: young and older adults.Methods: For the humans, 3-D computational models were created using wall property data from the literature as well as patient-specific magnetic resonance imaging (MRI) and non-invasive hemodynamic data; for the mice, comparable models were created using population-based properties and hemodynamics as well as subject-specific anatomies. Global aortic hemodynamics and wall stiffness were compared between humans and mice across age groups.Results: For young adult subjects, we found differences between species in pulse pressure amplification, compliance and resistance distribution, and aortic stiffness gradient. We also found differences in response to aging between species. Generally, the human spatial gradients of stiffness and pulse pressure across the aorta diminished with age, while they increased for the mice.Conclusion: These results highlight key differences in vascular aging between human and mice, and it is important to acknowledge these when using mouse models for cardiovascular research.

show abstract

“…A pre-intervention 3D model of the Fontan pathway, including the IVC, SVC, LPA, RPA, and fenestration to the RA was constructed ( Figure 2A ) using the open-source cardiovascular simulation software CRIMSON ( Arthurs et al, 2021 ). Lumen center lines, and 2D segmentations of the vessel lumen ( Figure 2B ) were used to create a computer aided design (CAD) model of the cardiovascular anatomy using lofting and blending operations ( Figure 2C ).…”

Section: Methodsmentioning

confidence: 99%

Interventional Planning for Endovascular Revision of a Lateral Tunnel Fontan: A Patient-Specific Computational Analysis

et al. 2021

Self Cite

View full text Add to dashboard Cite

IntroductionA 2-year-old female with hypoplastic left heart syndrome (HLHS)-variant, a complex congenital heart defect (CHD) characterized by the underdevelopment of the left ventricle, presented with complications following single ventricle palliation. Diagnostic work-up revealed elevated Fontan pathway pressures, as well as significant dilation of the inferior Fontan pathway with inefficient swirling flow and hepatic venous reflux. Due to the frail condition of the patient, the clinical team considered an endovascular revision of the Fontan pathway. In this work, we performed a computational fluid dynamics (CFD) analysis informed by data on anatomy, flow, and pressure to investigate the hemodynamic effect of the endovascular Fontan revision.MethodsA patient-specific anatomical model of the Fontan pathway was constructed from magnetic resonance imaging (MRI) data using the cardiovascular modeling software CardiovasculaR Integrated Modeling and SimulatiON (CRIMSON). We first created and calibrated a pre-intervention 3D-0D multi-scale model of the patient’s circulation using fluid-structure interaction (FSI) analyses and custom lumped parameter models (LPMs), including the Fontan pathway, the single ventricle, arterial and venous systemic, and pulmonary circulations. Model parameters were iteratively tuned until simulation results matched clinical data on flow and pressure. Following calibration of the pre-intervention model, a custom bifurcated endograft was introduced into the anatomical model to virtually assess post-intervention hemodynamics.ResultsThe pre-intervention model successfully reproduced the clinical hemodynamic data on regional flow splits, pressures, and hepatic venous reflux. The proposed endovascular repair model revealed increases of mean and pulse pressure at the inferior vena cava (IVC) of 6 and 29%, respectively. Inflows at the superior vena cava (SVC) and IVC were each reduced by 5%, whereas outflows at the left pulmonary artery (LPA) and right pulmonary artery (RPA) increased by 4%. Hepatic venous reflux increased by 6%.ConclusionOur computational analysis indicated that the proposed endovascular revision would lead to unfavorable hemodynamic conditions. For these reasons, the clinical team decided to forgo the proposed endovascular repair and to reassess the management of this patient. This study confirms the relevance of CFD modeling as a beneficial tool in surgical planning for single ventricle CHD patients.

show abstract

CRIMSON: An open-source software framework for cardiovascular integrated modelling and simulation

Cited by 50 publications

References 56 publications

A Combined Computational Fluid Dynamics and Arterial Spin Labeling MRI Modeling Strategy to Quantify Patient-Specific Cerebral Hemodynamics in Cerebrovascular Occlusive Disease

A Combined Computational Fluid Dynamics and Arterial Spin Labeling MRI Modeling Strategy to Quantify Patient-Specific Cerebral Hemodynamics in Cerebrovascular Occlusive Disease

Comparative Study of Human and Murine Aortic Biomechanics and Hemodynamics in Vascular Aging

Interventional Planning for Endovascular Revision of a Lateral Tunnel Fontan: A Patient-Specific Computational Analysis

Contact Info

Product

Resources

About