• Home
• Preliminary Program

11:30   Vascular - II Chair: Frank Gijsen 11:30 15 mins Impact of catheter movement during Transarterial Radioembolization Treatment, an in vitro analysis Tess Snoeijink, Jan van der Hoek, Hadi Mirgolbabaee, Frank Nijsen, Erik Groot Jebbink Abstract: Introduction: Transarterial radioembolization (TARE) has become an established treatment method for primary and secondary liver cancer [1]. TARE consists of a catheter-based intraarterial infusion of radioactive microspheres. The microspheres are transported through the arterial liver vasculature until they lodge predominantly in and around the tumour to deliver a high local radiation dose [2]. Even though TARE has been used in clinical practice for over 20 years, the exact mechanisms behind the distribution remain unknown. Recent in silico studies point towards a major influence of axial catheter location on the distribution of microspheres [3-5]; however, it often lacks validation. The aim of the current study was to investigate the influence of axial catheter location and movement on the microsphere distribution in an idealized in vitro model. Method: An idealized, 2D, symmetrical, right hepatic artery phantom was developed which bifurcates three times into eight outlets. Hemodynamics were recreated using a programmable piston pump (SuperPump, ViVitro labs, Victoria, Canada). In total, 30 injections with non-irradiated holmium loaded poly(L-lactic acid) microspheres (Quirem Medical B.V., Deventer, The Netherlands) were performed with two types of catheters (Progreat 2.7F microcatheter and a rigid counterpart) at 0, 1.5 and 3.0 cm upstream of the first bifurcation. The outflow distribution was analysed by collecting the injected microspheres at each outlet and counting particle volumes using a Coulter Counter Multisizer 3 (Beckman Coulter Nederland, Mijdrecht, the Netherlands). Motion analysis over time and initial positioning of the catheter was performed using two cameras (top and side view, Logitech BRIO webcam, Logitech). Results: A larger longitudinal distance from the bifurcation led to a more homogeneous microsphere distribution among the outlets. For the clinical catheter, maximal deflections of 0.8 mm (top view) and 0.6 mm (side view) through the vessel lumen were observed, while the rigid catheter showed no movement. Conclusion: The current work confirms the influence of longitudinal catheter location on the distribution of microspheres. Moreover, movement of the clinical catheter through the lumen influenced the downstream distribution of microspheres, which should be further investigated. Future research should focus on whether the observed movement of the catheter also occurs in clinical practice. References: [1] P Hilgard et al., Hepatology 52 (5), 1741 (2010). [2] MTM Reinders et al., Seminars in Nuclear Medicine 49 (3), 237 (2019). [3] T Bomberna et al., Expert Opin Drug Deliv 18 (3), 409 (2021). [4] J Aramburu et al., J Biomech 49 (15), 3705 (2016). [5] A Taebi et al., J Biomech Eng 143 (1), 01 (2021). 11:45 15 mins Using a tissue-engineered model to investigate the impact of collagen orientation on the local mechanical behavior of atherosclerotic plaque caps Hanneke Crielaard, Tamar Wissing, Su Guvenir Torun, Pablo de Miguel Muñoz, Gert-Jan Kremers, Frank Gijsen, Ali Akyildiz, Kim van der Heiden Abstract: Stroke is commonly initiated by rupture of the atherosclerotic plaque fibrous cap in a carotid artery. However, cap rupture mechanisms are not well understood yet. Understanding the impact of structural components of the cap on its local mechanics may provide critical insights into plaque rupture. We tissue engineered collagenous plaque analogs [1], to scrutinize the reciprocal relationships between composition and mechanics. We aim to unravel plaque rupture mechanics by studying local mechanical properties and collagen orientation in a tissue-engineered plaque model. Ten collagenous cap analogs with a soft inclusion (SI), mimicking the plaque lipid core, were created [1]. Analogs were exposed to multiphoton microscopy with second harmonic generation to obtain local fiber orientation, using a fiber orientation analysis tool (FibLab). After imaging, analogs were exposed to uniaxial tensile tests until full rupture. Local (Green-Lagrange) strains under tensile stretching were measured through DIC analysis using the software Ncorr [2]. Collagen fibers were mainly oriented in the circumferential direction (y-direction). In the center of the samples, larger deviations in fiber orientation from the y- direction and a more dispersed fiber architecture were found compared to the edges. Rupture was found to initiate near high tensile strain regions in the soft inclusion (mimicking the lipid core), suggesting that rupture initiation does not always occur at the luminal surface of an atherosclerotic plaque, as is generally assumed. [1] Wissing TB, Van der Heiden K, Serra SM, Smits AIPM, Bouten CV, Gijsen FJH . Tissue-engineered collagenous fibrous cap models to systematically elucidate atherosclerotic plaque rupture. Scientific Reports 12(1), (2022) [2]. Blaber J, Adair B, Antoniou A, Ncorr: Open-Source 2D Digital Image Correlation Matlab Software. Experimental Mechanics, 55(6), 1105–1122 (2015). 12:00 15 mins Equivariant wall shear stress estimation on the Coronary artery wall Julian Suk Abstract: Computational fluid dynamics (CFD) is used for the non-invasive estimation of hemodynamic biomarkers like wall shear stress (WSS) from 3D models of human arteries. However, long computation times limit the widespread applicability of CFD in clinical practice. As an alternative, we propose to use a graph convolutional network (GCN) operating on a surface mesh. Our GCN can produce accurate directional WSS estimates mapped to the mesh vertices in under five seconds [1]. For training, we randomly generate datasets of single and bifurcating coronary artery models and simulate steady and pulsatile blood flow, subject to varying boundary conditions. Surface meshes with WSS vectors mapped to their vertices are used to train a GCN consisting of gauge-equivariant mesh (GEM) convolution [2] and pooling layers. The network is provably SE(3)-equivariant: shifting the input mesh in space does not affect WSS vectors, while a rotation results in accordingly rotated WSS vectors. GEM-GCN can predict WSS fields jointly for a set of discrete time points in the cardiac cycle. We condition the network on simulation-specific coronary blood flow by appending it as scalar field to the input features. In previously unseen arteries, our neural network produces WSS estimations with an approximation error of 7.8 % and 11.9 % for the single and bifurcating arteries, respectively. Predicting directional WSS on a surface mesh takes less than 5 seconds, compared to more than ten minutes for the corresponding CFD simulation. Our results indicate that GEM-GCN has the potential to be a feasible surrogate for CFD in time-critical applications. Further research is needed to assess generalisation to patient- specific 3D artery models. References [1] Suk, J., de Haan, P., Lippe, P., Brune, C., and Wolterink, J.M. (2021). Mesh convolutional neural networks for wall shear stress estimation in 3d artery models. In Statistical Atlases and Computational Models of the Heart. [2] de Haan, P., Weiler, M., Cohen, T., and Welling, M. (2021). Gauge equivariant mesh CNNs: anisotropic convolutions on geometric graphs. In Proceedings of the 9th International Conference on Learning Representations. 12:15 15 mins Ultrasound-based fluid-structure interaction modeling of Abdominal Aortic Aneurysms: model complexity and personalization Judith Fonken, Eline van Engelen, Esther Maas, Arjet Nievergeld, Marc van Sambeek, Frans van de Vosse, Richard Lopata Abstract: Insight in abdominal aortic aneurysm (AAA) development, growth and rupture risk requires a large, longitudinal study on mechanical properties of AAAs. For patient-specific risk assessment and analysis of the mechanical state, fluid-structure interaction (FSI) models are considered, which require the AAA geometry and its dynamics. Time-resolved 3-dimensional ultrasound (3D+t US) is the preferred image modality to extract the patient-specific geometry, since it is safe, fast, affordable and provides functional information such as wall motion and blood velocity. A previous study has shown the feasibility of 3D+t US-based FSI simulations [1]. In this study, the FSI framework was improved vastly to better approach the in-vivo situation. Due to the limited field-of-view of 3D+t US, the aorto-iliac bifurcation geometry is often not included in the US acquisition. Due to this limitation, a single outlet AAA geometry was used in our previous study [1]. However, the bifurcation geometry does influence the hemodynamics in the aneurysm region. This study showed the feasibility of adding a parametric bifurcation to the aneurysm geometry, with median differences in hemodynamics below 1% with respect to the patient-specific bifurcation geometry. Furthermore, the framework is further personalized by including patient-specific flow parameters derived from US Doppler acquisitions. These flow parameters include the flow pulse, velocity profile over the vessel cross-section, inlet radius and inlet distance. FSI simulations employing patient-specific flow parameters yielded mean differences of 168% (TAWSS), 40% (OSI) and 7% (wall stress), with respect to simulations using generic flow parameters. Finally, the AAA geometry was embedded in a surrounding soft tissue and spine was included at the posterior side modelled by a stiff rod. This resulted in a decrease (mean: 34%) in displacement, especially at the posterior wall, and a homogenization and decrease (mean: 14%) in wall stress, similar to [2]. In future studies, the obtained FSI framework will be further personalized using 4D US speckle tracking for wall motion [3], and validated with the use of 4D flow MRI. The envisioned framework for realistic and personalized 3D+t US-based FSI simulations paves the way for longitudinal studies on AAA development, growth, and rupture risk. [1] Fonken et al., Front. Physiol., 1255, 2021. [2] Petterson et al., Journal of Biomechanics, 126-133, 2019 [3] Disseldorp et al., Eur. Heart J. Cardiovasc. Imaging., 185-191, 2019. 12:30 15 mins Constituent-based quasi-linear viscoelasticity: Capturing non-linear viscoelasticity with quasi-linear models Alessandro Giudici, Koen van der Laan, Myrthe van der Bruggen, Shaiv Parikh, Eline Berends, Sébastien Foulquier, Tammo Delhaas, Koen Reesink, Bart Spronck Abstract: Arteries exhibit complex viscoelastic behaviours, which are highly non-linear both in terms of elasticity and viscosity. Previous works have proposed different solutions to model elastic non-linearity, which capture well the arterial wall response to quasi-static deformations. However, in vivo the arterial wall is subjected to pulsatile loads for which viscoelastic phenomena cannot be neglected. Unlike for elasticity, effective solutions to model non-linear viscoelasticity are lacking. On the one hand, quasi-linear viscoelasticity (QLV) offers a practical solution to viscoelastic modelling, but its linear viscosity assumption is unsuitable for whole-wall vascular application. On the other hand, deformation-dependent parameters make fully non-linear viscoelastic models impractical. Indeed, their application to experimental data often leads to identifying specific solutions for each tested loading condition. In the present study, we address this issue by applying QLV theory at the wall constituent rather than at the whole-wall level. Five murine common carotid arteries were subjected to an experimental protocol of quasi-static and harmonic biaxial loading conditions for viscoelastic mechanical characterisation. In our constituent-based QLV (cbQLV) framework, the arterial wall was modelled as a constrained mixture of an isotropic elastin matrix and four families of collagen fibres in which collagen and elastin were assigned different stress relaxation functions. Non-linearity in viscoelasticity was quantified in terms of dependency of the dynamic-to-quasi-static stiffness ratio on pressure, and the performance of our model was compared to that of standard QLV (sQLV). The experimentally measured dynamic-to-quasi-static stiffness ratio was negligible at low pressures (1.03±0.03 at a pressure range 40–80 mmHg; mean±standard deviation) and rose with increasing pressure (1.58±0.22 at 120–160 mmHg, Figure 1A). By assigning viscoelastic behaviour to collagen and almost purely elastic behaviour to elastin, cbQLV captured well the pressure dependency of this ratio (Figure 1B). Conversely, sQLV failed to capture this complex behaviour (Figure 1C). In conclusion, constituent-based QLV offers a practical solution to model complex non-linear viscoelastic behaviours using a unique set of deformation-independent viscous parameters. 12:45 15 mins Personalized region-specific characterization of abdominal aortic aneurysms using 4D ultrasound and modified virtual fields method Mirunalini Thirugnanasambandam, Esther Maas, Arjet Nievergeld, Marc van Sambeek, Stephane Avril, Richard Lopata Abstract: Assessment of patient-specific behaviour of abdominal aortic aneurysms (AAA) is crucial to evaluating personalized biomechanical rupture risk indices with better accuracy. The ability of 4D ultrasound (US) to assess AAA wall motion, when combined with a novel inverse method, will provide an innovative framework for evaluation of individualized material behaviour of AAAs. 4D-US images were acquired from patients in a supine position over multiple cardiac cycles at an acquisition rate of 4-8 volumes/second. Image volumes were segmented at diastole, and a 3D speckle tracking algorithm was used to track the AAA walls to the systolic configuration. B-spline grids were fitted to the systolic and diastolic geometries of both walls. The displacement at each node of each wall was evaluated based on the minimization of the distance between the systolic and diastolic grids post-co-registration. Displacement of the nodes in the bulk of the AAA wall were evaluated using a weighted linear interpolation of the displacement vectors of their inner and outer wall neighbours. The smoothed 3D displacement field was input to the modified virtual field method (mVFM) [1], which iteratively evaluated the optimized material parameters based on a virtual work-based cost function. Thus, an automated framework for computationally inexpensive estimation of patient-specific material properties was formulated. The aforementioned technique was implemented, validated in-silico, and applied to multiple patient-specific AAAs. An uncoupled Neo-Hookean formulation was used to describe AAA wall material behaviour. With an initial guess of c10 = 1.24.105 Pa, optimal c10 values were predicted in three different regions (healthy abdominal aorta, anterior and posterior AAA sac) within 10 iterations using the 4D US+mVFM framework. The anterior AAA sac and the healthy abdominal aorta had the highest and lowest shear modulus, respectively. In the future, personalized material parameters will be evaluated while using sophisticated constitutive models. References: [1] Mei Y et al. J Elast, 145, 265-194 (2021).

end %-->