My Publications by Sarah Vigmostad
From medical images to flow computations without user-generated meshes
Biomedical flow computations in patient-specific geometries require integrating image acquisition... more Biomedical flow computations in patient-specific geometries require integrating image acquisition and processing with fluid flow solvers. Typically, image-based modeling processes involve several steps, such as image segmentation, surface mesh generation, volumetric flow mesh generation, and finally, computational simulation. These steps are performed separately, often using separate pieces of software, and each step requires considerable expertise and investment of time on the part of the user. In this paper, an alternative framework is presented in which the entire image-based modeling process is performed on a Cartesian domain where the image is embedded within the domain as an implicit surface. Thus, the framework circumvents the need for generating surface meshes to fit complex geometries and subsequent creation of body-fitted flow meshes. Cartesian mesh pruning, local mesh refinement, and massive parallelization provide computational efficiency; the image-to-computation techniques adopted are chosen to be suitable for distributed memory architectures. The complete framework is demonstrated with flow calculations computed in two 3D image reconstructions of geometrically dissimilar intracranial aneurysms. The flow calculations are performed on multiprocessor computer architectures and are compared against calculations performed with a standard multistep route. Copyright (c) 2014 John Wiley & Sons, Ltd.
Patient-specific bicuspid valve dynamics: Overview of methods and challenges
About 1-2 % of the babies are born with bicuspid aortic valves instead of the normal aortic valve... more About 1-2 % of the babies are born with bicuspid aortic valves instead of the normal aortic valve with three leaflets. A significant portion of the patients with the congenital bicuspid valve morphology suffer from aortic valve stenosis and/or ascending aortic dilatation and dissection thus requiring surgical intervention when they are young adults. Patients with bicuspid aortic valves (BAVs) have also been found to develop valvular stenosis earlier than those with the normal aortic valve. This paper overviews current knowledge of BAVs, where several studies have suggested that the mechanical stresses induced on the valve leaflets and the abnormal flow development in the ascending aorta may be an important factor in the diseases Of the valve and the aortic root. The long-term goals of the studies being performed in our laboratory are aimed towards potential stratification of bicuspid valve patients who may be at risk for developing these pathologies based on analyzing the hemodynamic environment of these valves using fluid-structure interaction (FSI) modeling. Patient-specific geometry of the normal tri-cuspid and bicuspid valves are reconstructed from real-time 3D ultrasound images and the dynamic analyses performed in order to determine the potential effects of mechanical stresses on the valve leaflet and aortic root pathology. This paper describes the details of the computational tools and discusses challenges with patient-specific modeling. (C) 2012 Elsevier Ltd. All rights reserved.
Techniques to derive geometries for image-based Eulerian computations
Purpose - The performance of three frequently used level set-based segmentation methods is examin... more Purpose - The performance of three frequently used level set-based segmentation methods is examined for the purpose of defining features and boundary conditions for image-based Eulerian fluid and solid mechanics models. The focus of the evaluation is to identify an approach that produces the best geometric representation from a computational fluid/solid modeling point of view. In particular, extraction of geometries from a wide variety of imaging modalities and noise intensities, to supply to an immersed boundary approach, is targeted.
Design/methodology/approach - Two- and three-dimensional images, acquired from optical, X-ray CT, and ultrasound imaging modalities, are segmented with active contours, k-means, and adaptive clustering methods. Segmentation contours are converted to level sets and smoothed as necessary for use in fluid/solid simulations. Results produced by the three approaches are compared visually and with contrast ratio, signal-to-noise ratio, and contrast-to-noise ratio measures.
Findings - While the active contours method possesses built-in smoothing and regularization and produces continuous contours, the clustering methods (k-means and adaptive clustering) produce discrete (pixelated) contours that require smoothing using speckle-reducing anisotropic diffusion (SRAD). Thus, for images with high contrast and low to moderate noise, active contours are generally preferable. However, adaptive clustering is found to be far superior to the other two methods for images possessing high levels of noise and global intensity variations, due to its more sophisticated use of local pixel/voxel intensity statistics.
Originality/value - It is often difficult to know a priori which segmentation will perform best for a given image type, particularly when geometric modeling is the ultimate goal. This work offers insight to the algorithm selection process, as well as outlining a practical framework for generating useful geometric surfaces in an Eulerian setting.
MICRO-SCALE BLOOD PARTICULATE DYNAMICS USING A NURBS-BASED ISOGEOMETRIC ANALYSIS
The current research presents a novel method in which blood particulates – biconcave red blood ce... more The current research presents a novel method in which blood particulates – biconcave red blood cells (RBCs) and spherical cells are modeled using isogeometric analysis, specifically Non-Uniform Rational B-Splines(NURBS) in 3-D. The use of NURBS ensures that even with a coarse representation, the geometry of the blood particulates maintains an accurate description when subjected to large deformations. The fundamental advantage of this method is the coupling of the geometrical description and the stress analysis of the cell membrane into a single, unified framework. Details on the modeling approach, implementation of boundary conditions and the membrane mechanics analysis using isogeometric modeling are presented, along with validation cases for spherical and biconcave cells. Using NURBS – based isogeometric analysis, the behavior of individual cells in fluid flow is presented and analyzed in different flow regimes using as few as 176 elements for a spherical cell and 220 elements for a biconcave RBC. This work provides a framework for modeling a large number of 3-D deformable biological cells, each with its own geometric description and membrane properties. To the best knowledge of the authors, this is the first application of NURBS – based isogeometric analysis to simulate blood particulates in flow in 3-D
A Multiwell Disc Appliance Used to Deliver Quantifiable Accelerations and Shear Stresses at Sonic Frequencies
There are numerous examples of fluid–structure interactions (FSIs) within the human body. In all ... more There are numerous examples of fluid–structure interactions (FSIs) within the human body. In all cases, a computer model capable of simulating the phenomenon can aid in the understanding of organ function, failure, and implant design or improvement. In the current paper, two approaches are examined for use in simulating the FSI problem of the dynamics of tissue heart valves. Valve leaflets have nonlinear anisotropic material properties, and undergo complex deformation. Their motion affects—and is affected by—the surrounding blood. This two-way coupling necessitates a robust algorithm in order to overcome numerical stiffness, convergence challenges, and stability issues. A locally refined Cartesian mesh, sharp interface method has been developed for the fluid flow solution. In the structural domain, the valve leaflet is represented in a Lagrangian fashion and moves based on its experimentally determined material properties. In computing leaflet motion, the anisotropic, nonlinear material properties of the valve leaflet are incorporated using a finite element solver, which calculates the leaflet deformation and stresses based on the stress imparted by the surrounding fluid. Two FSI algorithms have been studied in the context of a sharp-interface Cartesian grid setting, and each has been validated with benchmark results. The two approaches are compared, and ultimately one is selected as most appropriate for simulating tissue heart valves. In the selected approach, a strongly coupled, partitioned method is used in which subiterations of the fluid and structure solutions are performed at each time step. During the subiterations, the leaflet motion is used as a boundary condition on the fluid, and the fluid stresses act as a boundary condition on the leaflet. In this way, continuity is ensured and two-way coupling is achieved. The selected approach has overcome the challenges faced by previous simulations reported in the literature, and a robust FSI solution is achieved using physiologic Reynolds numbers, realistic material properties, highly resolved grids, and a dynamic simulation. This approach has the advantage of handling both thin and volumetric embedded objects in a unified fashion, and of treating rigid and deformable structures in the same way, thus allowing a spectrum of potential applications.
The mechanisms of plaque development in coronary arteries are not yet completely understood. Vess... more The mechanisms of plaque development in coronary arteries are not yet completely understood. Vessel geometry influences the local hemodynamics within a vessel, and the resulting wall shear stress in turn influences plaque development. Previously, we showed in-vivo that plaque tends to accumulate more on the inner curvature of a vessel than on its outer curvature. While vessel curvature is preserved during plaque progression, the local wall shear stresses change with lumen narrowing. The aim of this study was to test how the hypothesis that locations of low wall shear stress coincide with circumferentially larger plaque accumulation depends on vascular remodeling with or without lumen narrowing. We have analyzed 39 in-vivo intravascular-ultrasound pullbacks, for which geometrically accurate 3-D models were obtained by fusion with x-ray angiography. Distorting subsegments (branches, calcifications, stents) were discarded, and the relative number of vessel locations was determined within a 10-40% area-stenosis range. This range corresponds to compensatory enlargement (outward or positive vessel remodeling), but not yet lumen narrowing, and these vessel segments were a focus of our study. For each segment, we determined the relative number of vessel locations for which circumferentially low wall shear stress coincided with larger plaque thickness and vice versa. The inverse association between wall shear stress and plaque thickness was significantly more pronounced (p<0.005) in vessel cross sections exhibiting compensatory enlargement without luminal narrowing than when the full spectrum of vessel stenosis severity was considered. Thus, the hypothesis is supported more in subsegments with less developed disease.
Comparison of Left Anterior Descending Coronary Artery Hemodynamics Before and After Angioplasty
Journal of Biomechanical Engineering-transactions of The Asme, 2006
Coronary artery disease (CAD) is characterized by the progression of atherosclerosis, a complex p... more Coronary artery disease (CAD) is characterized by the progression of atherosclerosis, a complex pathological process involving the initiation, deposition, development, and breakdown of the plaque. The blood flow mechanics in arteries play a critical role in the targeted locations and progression of atherosclerotic plaque. In coronary arteries with motion during the cardiac contraction and relaxation, the hemodynamic flow field is substantially different from the other arterial sites with predilection of atherosclerosis. In this study, our efforts focused on the effects of arterial motion and local geometry on the hemodynamics of a left anterior descending (LAD) coronary artery before and after clinical intervention to treat the disease. Three-dimensional (3D) arterial segments were reconstructed at 10 phases of the cardiac cycle for both pre- and postintervention based on the fusion of intravascular ultrasound (IVUS) and biplane angiographic images. An arbitrary Lagrangian-Eulerian formulation was used for the computational fluid dynamic analysis. The measured arterial translation was observed to be larger during systole after intervention and more out-of-plane motion was observed before intervention, indicating substantial alterations in the cardiac contraction after angioplasty. The time averaged axial wall shear stress ranged from -0.2 to 9.5 Pa before intervention compared to -0.02 to 3.53 Pa after intervention. Substantial oscillatory shear stress was present in the preintervention flow dynamics compared to that in the postintervention case.
Analysis of the Interdependencies Among Plaque Development, Vessel Curvature, and Wall Shear Stress in Coronary Arteries
The relationships among vascular geometry, hemodynamics, and plaque development in coronary arter... more The relationships among vascular geometry, hemodynamics, and plaque development in coronary arteries are not yet well understood. This in-vivo study was based on the observation that plaque frequently develops at the inner curvature of a vessel, presumably due to a relatively lower wall shear stress. We have shown that circumferential plaque distribution depends on the vessel curvature in the majority of vessels. Consequently, we studied the correlation of plaque distribution and hemodynamics in a set of 48 vessel segments reconstructed by 3-D fusion of intravascular ultrasound and x-ray angiography. The inverse relationship between local wall shear stress and plaque thickness was significantly more pronounced (p
A Novel Flex-Stretch-Flow Bioreactor for the Study of Engineered Heart Valve Tissue Mechanobiology
Annals of Biomedical Engineering, 2008
Tissue engineered heart valves (TEHV) have been observed to respond to mechanical conditioning in... more Tissue engineered heart valves (TEHV) have been observed to respond to mechanical conditioning in vitro by expression of activated myofibroblast phenotypes followed by improvements in tissue maturation. In separate studies, cyclic flexure, stretch, and flow (FSF) have been demonstrated to exhibit both independent and coupled stimulatory effects. Synthesis of these observations into a rational framework for TEHV mechanical conditioning has been limited, however, due to the functional complexity of tri-leaflet valves and the inherent differences of separate bioreactor systems. Toward quantifying the effects of individual mechanical stimuli similar to those that occur during normal valve function, a novel bioreactor was developed in which FSF mechanical stimuli can be applied to engineered heart valve tissues independently or in combination. The FSF bioreactor consists of two identically equipped chambers, each having the capacity to hold up to 12 rectangular tissue specimens (25 × 7.5 × 1 mm) via a novel “spiral-bound” technique. Specimens can be subjected to changes-in-curvature up to 50 mm−1 and uniaxial tensile strains up to 75%. Steady laminar flow can be applied by a magnetically coupled paddlewheel system. Computational fluid dynamic (CFD) simulations were conducted and experimentally validated by particle image velocimetry (PIV). Tissue specimen wall shear stress profiles were predicted as a function of paddlewheel speed, culture medium viscosity, and the quasi-static state of specimen deformation (i.e., either undeformed or completely flexed). Velocity profiles predicted by 2D CFD simulations of the paddlewheel mechanism compared well with PIV measurements, and were used to determine boundary conditions in localized 3D simulations. For undeformed specimens, predicted inter-specimen variations in wall shear stress were on average ±7%, with an average wall shear stress of 1.145 dyne/cm2 predicted at a paddlewheel speed of 2000 rpm and standard culture conditions. In contrast, while the average wall shear stress predicted for specimens in the quasi-static flexed state was ∼59% higher (1.821 dyne/cm2), flexed specimens exhibited a broad intra-specimen wall shear stress distribution between the convex and concave sides that correlated with specimen curvature, with peak wall shear stresses of ∼10 dyne/cm2. This result suggests that by utilizing simple flexed geometric configurations, the present system can also be used to study the effects of spatially varying shear stresses. We conclude that the present design provides a robust tool for the study of mechanical stimuli on in vitro engineered heart valve tissue formation.
Fluid Dynamic Analysis in a Human Left Anterior Descending Coronary Artery with Arterial Motion
Annals of Biomedical Engineering, 2004
A computational fluid dynamic (CFD) analysis is presented to describe local flow dynamics in both... more A computational fluid dynamic (CFD) analysis is presented to describe local flow dynamics in both 3-D spatial and 4-D spatial and temporal domains from reconstructions of intravascular ultrasound (IVUS) and bi-plane angiographic fusion images. A left anterior descending (LAD) coronary artery segment geometry was accurately reconstructed and subsequently its motion was incorporated into the CFD model. The results indicate that the incorporation of motion had appreciable effects on blood flow patterns. The velocity profiles in the region of a stenosis and the circumferential distribution of the axial wall shear stress (WSS) patterns in the vessel are altered with the wall motion introduced in the simulation. The time-averaged axial WSS between simulations of steady flow and unsteady flow without arterial motion were comparable (−0.3 to 13.7 Pa in unsteady flow versus −0.2 to 10.1 Pa in steady flow) while the magnitudes decreased when motion was introduced (0.3–4.5 Pa). The arterial wall motion affects the time-mean WSS and the oscillatory shear index in the coronary vessel fluid dynamics and may provide more realistic predictions on the progression of atherosclerotic disease.
A common hypothesis is that plaque accumulation in curved vessels is biased towards the inner ben... more A common hypothesis is that plaque accumulation in curved vessels is biased towards the inner bend of the curvature rather than the outer bend of the curvature. This bias in circumferential plaque distribution is likely associated with lower wall shear stress on the inner bend of the curved vessel. We quantitatively analyzed this effect in a set of 37 in-vivo human coronary artery segments from 31 patients. Three-dimensional models of the arteries were generated by an established system for fusion of image data from Xray angiography and intravascular ultrasound. Our results showed that the hypothesis held in the majority of vessels (p < 0.001), and that the effect increases with curvature. However, no evidence could be found for a direct relationship between plaque distribution and curvature in complex vessel geometries, thus motivating a more detailed analysis of wall shear stress patterns and their impact on circumferential plaque distribution.
Plaque development, vessel curvature, and wall shear stress in coronary arteries assessed by X-ray angiography and intravascular ultrasound
Medical Image Analysis, 2006
The relationships among vascular geometry, hemodynamics, and plaque development in the coronary a... more The relationships among vascular geometry, hemodynamics, and plaque development in the coronary arteries are complex and not yet well understood. This paper reports a methodology for the quantitative analysis of in vivo coronary morphology and hemodynamics, with particular emphasis placed on the critical issues of image segmentation and the automated classification of disease severity. We were motivated by the observation that plaque more often developed at the inner curvature of a vessel, presumably due to the relatively lower wall shear stress at these locations. The presented studies are based on our validated methodology for the three-dimensional fusion of intravascular ultrasound (IVUS) and X-ray angiography, introducing a novel approach for IVUS segmentation that incorporates a robust, knowledge-based cost function and a fully optimal, three-dimensional segmentation algorithm. Our first study shows that circumferential plaque distribution depends on local vessel curvature in the majority of vessels. The second study analyzes the correlation between plaque distribution and wall shear stress in a set of 48 in vivo vessel segments. The results were conclusive for both studies, with a stronger correlation of circumferential plaque thickness with local curvature than with wall shear stress. The inverse relationship between local wall shear stress and plaque thickness was significantly more pronounced (p&amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;amp;lt;0.025) in vessel cross sections exhibiting compensatory enlargement (positive remodeling) without luminal narrowing than when the full spectrum of disease severity was considered. The inverse relationship was no longer observed in vessels where less than 35% of vessel cross sections remained without luminal narrowing. The findings of this study confirm, in vivo, the hypothesis that relatively lower wall shear stress is associated with early plaque development.
Multidimensional segmentation of coronary intravascular ultrasound images using knowledge-based methods
In vivo studies of the relationships that exist among vascular geometry, plaque morphology, and h... more In vivo studies of the relationships that exist among vascular geometry, plaque morphology, and hemodynamics have recently been made possible through the development of a system that accurately reconstructs coronary arteries imaged by x-ray angiography and intravascular ultrasound (IVUS) in three dimensions. Currently, the bottleneck of the system is the segmentation of the IVUS images. It is well known that IVUS images contain numerous artifacts from various sources. Previous attempts to create automated IVUS segmentation systems have suffered from either a cost function that does not include enough information, or from a non-optimal segmentation algorithm. The approach presented in this paper seeks to strengthen both of those weaknesses -- first by building a robust, knowledge-based cost function, and then by using a fully optimal, three-dimensional segmentation algorithm. The cost function contains three categories of information: a compendium of learned border patterns, information theoretic and statistical properties related to the imaging physics, and local image features. By combining these criteria in an optimal way, weaknesses associated with cost functions that only try to optimize a single criterion are minimized. This cost function is then used as the input to a fully optimal, three-dimensional, graph search-based segmentation algorithm. The resulting system has been validated against a set of manually traced IVUS image sets. Results did not show any bias, with a mean unsigned luminal border positioning error of 0.180 +/- 0.027 mm and an adventitial border positioning error of 0.200 +/- 0.069 mm.
A Fluid-Structure Interaction Model for Artificial Tissue Heart Valves Using a Sharp Interface Fixed Grid Method
A tissue heart valve can be described as a deformable hyperelastic structure surrounded by a visc... more A tissue heart valve can be described as a deformable hyperelastic structure surrounded by a viscous fluid. The modeling of this complex system requires a fluid-structure interaction approach. A sharp interface, fixed Cartesian grid method is presented which accurately computes the motion and stresses of both the valve and the surrounding fluid. The fluid-structure interaction model couples the fluid and leaflet motions and stresses. This model has been developed to incorporate the normal and shear stresses developed in the leaflet as jumps in the pressure and shear stresses of the surrounding fluid. Stresses in the leaflet result from deformation, where the motion of the leaflet takes into account its experimentally derived material properties. A finite element solver calculates the leaflet deformation and stresses based on the conditions of surrounding fluid. Validations of the fluid-structure interaction model have been performed, and this method is currently being extended to three dimensions.
Journal of Biomechanics, 2006
was selected to describe the nonlinear material properties where parameters were chosen from publ... more was selected to describe the nonlinear material properties where parameters were chosen from published data. Results show that (1) the study is able to predict stress distributions on each components of plaque; (2) high stress more likely occurs at the shoulder regions of plaque, especially where the fibrous caps are thin; and (3) for the entire plaque, larger differences of stress and displacement between the planar and non-planar models can be found on the plaque regions near carotid bulb where the out of plane bending started.
Journal of Voice, 2011
Elastic characteristics of the pig, sheep and cow vocal folds were investigated through a series ... more Elastic characteristics of the pig, sheep and cow vocal folds were investigated through a series of in vitro experiments. Sample strips of the vocal fold tissue were dissected from pig, sheep and cow vocal folds and mounted inside a saline-filled ergometer chamber that was maintained at 37°C ± 1°C. Sinusoidal elongation was applied on the samples to obtain the passive force measurements. Force and elongation data from the samples were recorded electronically with a dual-servo system (ergometer). Stress-Strain data were compared to characterize the interspecies differences in the elastic properties of vocal folds. Pig vocal folds exhibited the most nonlinear stress-strain relationship, indicating the presence of a high level of collagen fibers. Cow vocal folds had the highest Young's modulus, but the tissue displayed a nearly linear stress-strain profile. Previous studies of phonation in these three species have indicated that pig larynges have the highest range of phonation frequencies, making them a good candidate for animal studies. The current study provides quantitative data for the elastic properties of the oscillating laryngeal tissue in these species and indicates that nonlinear behavior of these tissues may lead to wider oscillation ranges.
Measurement of Vocal Folds Elastic Properties for Continuum Modeling
Red Blood Cell Flow in the Cardiovascular System: A Fluid Dynamics Perspective
Peak Wall Stress Predicts Expansion Rate in Descending Thoracic Aortic Aneurysms
Uploads
My Publications by Sarah Vigmostad
Design/methodology/approach - Two- and three-dimensional images, acquired from optical, X-ray CT, and ultrasound imaging modalities, are segmented with active contours, k-means, and adaptive clustering methods. Segmentation contours are converted to level sets and smoothed as necessary for use in fluid/solid simulations. Results produced by the three approaches are compared visually and with contrast ratio, signal-to-noise ratio, and contrast-to-noise ratio measures.
Findings - While the active contours method possesses built-in smoothing and regularization and produces continuous contours, the clustering methods (k-means and adaptive clustering) produce discrete (pixelated) contours that require smoothing using speckle-reducing anisotropic diffusion (SRAD). Thus, for images with high contrast and low to moderate noise, active contours are generally preferable. However, adaptive clustering is found to be far superior to the other two methods for images possessing high levels of noise and global intensity variations, due to its more sophisticated use of local pixel/voxel intensity statistics.
Originality/value - It is often difficult to know a priori which segmentation will perform best for a given image type, particularly when geometric modeling is the ultimate goal. This work offers insight to the algorithm selection process, as well as outlining a practical framework for generating useful geometric surfaces in an Eulerian setting.