Elena S. Di Martino

Three-Dimensional Geometrical Characterization of Abdominal Aortic Aneurysms: Image-Based Wall Thickness Distribution

The clinical assessment of abdominal aortic aneurysm (AAA) rupture risk is based on the quantification of AAA size by measuring its maximum diameter from computed tomography (CT) images and estimating the expansion rate of the aneurysm sac over time. Recent findings have shown that geometrical shape and size, as well as local wall thickness may be related to this risk; thus, reliable noninvasive image-based methods to evaluate AAA geometry have a potential to become valuable clinical tools. Utilizing existing CT data, the three-dimensional geometry of nine unruptured human AAAs was reconstructed and characterized quantitatively. We propose and evaluate a series of 1D size, 2D shape, 3D size, 3D shape, and second-order curvature-based indices to quantify AAA geometry, as well as the geometry of a size-matched idealized fusiform aneurysm and a patient-specific normal abdominal aorta used as controls. The wall thickness estimation algorithm, validated in our previous work, is tested against discrete point measurements taken from a cadaver tissue model, yielding an average relative difference in AAA wall thickness of 7.8%. It is unlikely that any one of the proposed geometrical indices alone would be a reliable index of rupture risk or a threshold for elective repair. Rather, the complete geometry and a positive correlation of a set of indices should be considered to assess the potential for rupture. With this quantitative parameter assessment, future research can be directed toward statistical analyses correlating the numerical values of these parameters with the risk of aneurysm rupture or intervention (surgical or endovascular). While this work does not provide direct insight into the possible clinical use of the geometric parameters, we believe it provides the foundation necessary for future efforts in that direction.

Biaxial mechanical modeling of the small intestine.

Capsule endoscopes are pill-size devices provided with a camera that capture images of the small intestine from inside the body after being ingested by a patient. The interaction between intestinal tissue and capsule endoscopes needs to be investigated to optimize capsule design while preventing tissue damage. To that purpose, a constitutive model that can reliably predict the mechanical response of the intestinal tissue under complex mechanical loading is required. This paper describes the development and numerical validation of a phenomenological constitutive model for the porcine duodenum, jejunum and ileum. Parameters characterizing the mechanical behavior of the material were estimated from planar biaxial test data, where intestinal tissue specimens were simultaneously loaded along the circumferential and longitudinal directions. Specimen-specific Fung constitutive models were able to accurately predict the planar stress-strain behavior of the tested samples under a wide range of loading conditions. To increase model generality, average anisotropic constitutive relationships were also generated for each tissue region by fitting average stress-strain curves to the Fung potential. Due to the observed variability in the direction of maximum stiffness, the average Fung models were less anisotropic than the specimen-specific models. Hence, average isotropic models in the Neo-Hookean and Mooney-Rivlin forms were attempted, but they could not adequately describe the degree of nonlinearity in the tissue. Values of the R2 for the nonlinear regressions were 0.17, 0.44 and 0.93 for the average Neo-Hookean, Mooney-Rivlin and Fung models, respectively. Average models were successfully implemented into FORTRAN routines and used to simulate capsule deployment with a finite element method analysis.

Local Quantification of Wall Thickness and Intraluminal Thrombus Offer Insight into the Mechanical Properties of the Aneurysmal Aorta.

Wall stress is a powerful tool to assist clinical decisions in rupture risk assessment of abdominal aortic aneurysms. Key modeling assumptions that influence wall stress magnitude and distribution are the inclusion or exclusion of the intraluminal thrombus in the model and the assumption of a uniform wall thickness. We employed a combined numerical-experimental approach to test the hypothesis that abdominal aortic aneurysm (AAA) wall tissues with different thickness as well as wall tissues covered by different thrombus thickness, exhibit differences in the mechanical behavior. Ultimate tissue strength was measured from in vitro tensile testing of AAA specimens and material properties of the wall were estimated by fitting the results of the tensile tests to a histo-mechanical constitutive model. Results showed a decrease in tissue strength and collagen stiffness with increasing wall thickness, supporting the hypothesis of wall thickening being mediated by accumulation of non load-bearing components. Additionally, an increase in thrombus deposition resulted in a reduction of elastin content, collagen stiffness and tissue strength. Local wall thickness and thrombus coverage may be used as surrogate measures of local mechanical properties of the tissue, and therefore, are possible candidates to improve the specificity of AAA wall stress and rupture risk evaluations.

Is There a Role for Biomechanical Engineering in Helping to Elucidate the Risk Profile of the Thoracic Aorta

Clinical estimates of rupture and dissection risk of thoracic aortic aneurysms are based on nonsophisticated measurements of maximum diameter and growth rate. The use of aortic size alone may overlook the role that vessel heterogeneity plays in assessing the risk of catastrophic complications. Biomechanics may help provide a more nuanced approach to predict the behavior of thoracic aortic aneurysms. In this report, we review modeling studies with an emphasis on mechanical and fluid dynamics analyses. We identify open problems and highlight the future possibility of a multidisciplinary approach that includes biomechanics and imaging to evaluate the likelihood of rupture or dissection.

In vivo strain assessment of the abdominal aortic aneurysm

The only criteria currently used to inform surgical decision for abdominal aortic aneurysms are maximum diameter (>5.5 cm) and rate of growth, even though several studies have identified the need for more specific indicators of risk. Patient-specific biomechanical variables likely to affect rupture risk would be a valuable addition to the science of understanding rupture risk and prove to be a life saving benefit for patients. Local deformability of the aorta is related to the local mechanical properties of the wall and may provide indication on the state of weakening of the wall tissue. We propose a 3D image-based approach to compute aortic wall strain maps in vivo. The method is applicable to a variety of imaging modalities that provide sequential images at different phases in the cardiac cycle. We applied the method to a series of abdominal aneurysms imaged using cine-MRI obtaining strain maps at different phases in the cardiac cycle. These maps could be used to evaluate the distensibility of an aneurysm at baseline and at different follow-up times and provide an additional index to clinicians to facilitate decisions on the best course of action for a specific patient.

In vivo porcine left atrial wall stress: Computational model

Most computational models of the heart have so far concentrated on the study of the left ventricle, mainly using simplified geometries. The same approach cannot be adopted to model the left atrium, whose irregular shape does not allow morphological simplifications. In addition, the deformation of the left atrium during the cardiac cycle strongly depends on the interaction with its surrounding structures. We present a procedure to generate a comprehensive computational model of the left atrium, including physiological loads (blood pressure), boundary conditions (pericardium, pulmonary veins and mitral valve annulus movement) and mechanical properties based on planar biaxial experiments. The model was able to accurately reproduce the in vivo dynamics of the left atrium during the passive portion of the cardiac cycle. A shift in time between the peak pressure and the maximum displacement of the mitral valve annulus allows the appendage to inflate and bend towards the ventricle before the pulling effect associated with the ventricle contraction takes place. The ventricular systole creates room for further expansion of the appendage, which gets in close contact with the pericardium. The temporal evolution of the volume in the atrial cavity as predicted by the finite element simulation matches the volume changes obtained from CT scans. The stress field computed at each time point shows remarkable spatial heterogeneity. In particular, high stress concentration occurs along the appendage rim and in the region surrounding the pulmonary veins.

In vivo porcine left atrial wall stress: Effect of ventricular tachypacing on spatial and temporal stress distribution

Animal models of ventricular tachypacing (VTP) have been successfully used to reproduce the relevant features observed in patients with atrial fibrillation, such as increased atrial pressure and volume, ion-channel alterations and fibrosis. After performing VTP on a healthy Yorkshire pig, we measured an increase in volume of 60%, a two-fold rise in pressure, and a complex pattern of local mechanical, histological and biochemical changes, including a generalized stiffening of the wall. A protocol recently developed was employed to generate computational models of the porcine left atrium mechanics in healthy conditions and after VTP. Comparison of the stress distribution in the healthy vs. VTP case provided a map of how pressure overload affects and modifies left atrium mechanics. Overall, a positive increase in stress was computed after the VTP treatment. Regions of large increase in the stresses post-VTP were the appendage boundaries, the area around the lower pulmonary vein and the area in the front of the atrium towards the appendage. Due to the elevated stress, the back of the atrium mainly modified its mechanical response, while the appendage remodeled both its shape and its mechanical properties. Large changes in the shape of the mitral valve annulus could be observed as a consequence of the remodeling in the front of the atrium. The relation between local mechanical stress and remodeling that emerges from the results is in agreement with our hypothesis that the structural changes in the atrium are a consequence of a stress-mediated mechanism.

A Mechanical Characterization of the Porcine Atria at the Healthy Stage and After Ventricular Tachypacing

Atrial fibrillation (AF) is a cardiac arrhythmia that highly increases the risk of stroke and is associated with significant but still unexplored changes in the mechanical behavior of the tissue. Planar biaxial tests were performed on tissue specimens from pigs at the healthy stage and after ventricular tachypacing (VTP), a procedure applied to reproduce the relevant features of AF. The local arrangement of the fiber bundles in the tissue was investigated on specimens from rabbit atria by means of circularly polarized light. Based on this, mechanical data were fitted to two anisotropic constitutive relationships, including a four-parameter Fung-type model and a microstructurally-motivated model. Accounting for the fiber-induced anisotropy brought average R(2) = 0.807 for the microstructurally-motivated model and average R(2) = 0.949 for the Fung model. Validation of the fitted constitutive relationships was performed by means of FEM simulations coupled to FORTRAN routines. The performances of the two material models in predicting the second Piola-Kirchhoff stress were comparable, with average errors <3.1%. However, the Fung model outperformed the other in the prediction of the Green-Lagrange strain, with 9.2% maximum average error. To increase model generality, a proper averaging procedure accounting for nonlinearities was used to obtain average material parameters. In general, a stiffer behavior after VTP was noted.

Collagen structural alterations contribute to stiffening of tissue after split‐thickness skin grafting

The gold standard treatment for full thickness injuries of the skin is autologous split‐thickness skin grafting. This involves harvesting the epidermis and superficial dermis from healthy skin and transplanting it onto the prepared wound bed. The donor site regenerates spontaneously, but the appendages and cellular components from the dermal layer are excluded from the graft. As a result, the new tissue is inferior; the healed graft site is dry/itchy, has decreased elasticity, increased fragility, and altered sensory function. Because this dermal layer is composed of collagen and other extracellular matrix proteins, the aim was to characterize the changes in the dermal collagen after split thickness grafting that could contribute to a deficit in functionality. This will serve as a baseline for future studies designed to improve skin function using pharmacological or cell‐based therapies for skin repair. A xenograft model whereby human split‐thickness grafts were implanted into full‐thickness defects on immunocompromised (athymic Nu/Nu) mice was used. The grafts were harvested 4 and 8 weeks later. The collagen microstructure was assessed with second harmonic generation with dual‐photon microscopy and light polarization analysis. Collagen fiber stiffness and engagement stretch were estimated by fitting the results of biaxial mechanical tensile tests to a histo‐mechanical constitutive model. The stiffness of the collagen fibril–proteoglycan complex increased from 682 ± 226 kPa/sr to 1016 ± 324 kPa/sr between 4 and 8 weeks postgrafting. At the microstructural level there were significant decreases in both thickness of collagen fibers (3.60 ± 0.34 μm vs. 2.10 ± 0.27 μm) and waviness ratio (2.04 ± 0.17 vs. 1.43 ± 0.08) of the collagen fibers postgrafting. The decrease of the macroscopic engagement stretch from 1.19 ± 0.11 to 1.09 ± 0.08 over time postgrafting mirrored the decrease in waviness measured at the microscopic level. This suggested that the integrity of the collagen fibers was compromised and contributed to the functional deficit of the skin postgrafting.

A feature-based morphing methodology for computationally modeled biological structures applied to left atrial fiber directions.

To properly simulate the behavior of biological structures through computer modeling, there exists a need to describe parameters that vary locally. These parameters can be obtained either from literature or from experimental data and they are often assigned to regions in the model as lumped values. Furthermore, parameter values may be obtained on a representative case and may not be available for each specific modeled organ. We describe a semiautomated technique to assign detailed maps of local tissue properties to a computational model of a biological structure. We applied the method to the left atrium of the heart. The orientation of myocytes in the tissue as obtained from histologic analysis was transferred to the 3D model of a porcine left atrium. Finite element method (FEM) dynamic simulations were performed by using an isotropic, neo-Hookean, constitutive model first, then adding an anisotropic, cardiomyocyte oriented, Fung-type component. Results showed higher stresses for the anisotropic material model corresponding to lower stretches in the cardiomyocyte directions. The same methodology can be applied to transfer any map of parameters onto a discretized finite element model.