http://hdl.handle.net/10498/6774 Thu, 05 Nov 2026 12:00:31 GMT 2026-11-05T12:00:31Z http://hdl.handle.net/10498/40304 From tissue mechanics to clinical modelingA validated finite element framework of growth plate behavior based on mechanical experiments and patient data Hucke, Lucie The growth plate, a layer of hyaline cartilage between the epiphysis and metaphysis of long bones, governs longitudinal bone growth. A better understanding of its mechanical behavior can support treatments that intentionally modulate growth, such as guided growth therapy. However, current computational models often lack experimental validation and do not fully capture the nonlinear and time-dependent properties of the tissue. This thesis aimed to develop and validate a framework that links the experimental characterization of growth plate mechanics with finite element modeling. The main objectives were to (1) generate a finite element model to simulate guided growth loading, (2) experimentally determine the mechanical response of growth plate tissue under torsion and compression, (3) identify suitable constitutive models for its nonlinear and viscoelastic behavior, and (4) validate the resulting model using imaging-based strain data. A finite element model revealed that the implant applied locally high pressure stresses to the growth plate while reducing the cyclic loading, two factors that are reported in literature to inhibit long growth. Mechanical tests on porcine growth plates revealed a nonlinear and time-dependent behavior. The Yeoh hyperelastic model with three terms best represented the long-term response, and viscoelasticity was captured with a two-term Prony series. Histological analyzes indicated that local structural differences correspond to variations in stiffness. The experimentally derived parameters were validated using MRI-based strain mapping, showing a correspondence between the predicted and measured strain fields.; La placa de crecimiento, una capa de cartílago hialino entre la epífisis y la metáfisis, controla el crecimiento longitudinal de los huesos. Comprender su comportamiento mecánico puede apoyar terapias que modulan el crecimiento, como la guiada. Sin embargo, los modelos computacionales actuales carecen de validación experimental y no captan sus propiedades no lineales y dependientes del tiempo. El objetivo de esta tesis fue desarrollar y validar un marco que relacione la caracterización experimental de la mecánica de la placa de crecimiento con el modelado por elementos finitos. Los objetivos fueron (1) generar un modelo de elementos finitos para simular la carga de crecimiento guiado, (2) determinar experimentalmente la respuesta mecánica bajo torsión y compresión, (3) identificar modelos constitutivos adecuados, y (4) validar el modelo con datos de deformación basados en imágenes. El modelo reveló que el implante aplicaba tensiones de alta presión localmente a la placa, reduciendo la carga cíclica, ambos factores que inhiben el crecimiento. Las pruebas en placas porcinas mostraron un comportamiento no lineal y dependiente del tiempo. El modelo hiperelástico de Yeoh con tres términos describió mejor la respuesta, y la viscoelasticidad se capturó con una serie de Prony de dos términos. Los análisis histológicos vincularon variaciones estructurales con diferencias en rigidez, y la validación por resonancia magnética mostró correspondencia entre los campos de deformación previstos y medidos. Thu, 01 Jan 2026 00:00:00 GMT http://hdl.handle.net/10498/40304 2026-01-01T00:00:00Z http://hdl.handle.net/10498/39996 Development and Validation of Non-Invasive Methods for the Identification of Patient-Specific Hyperelastic and Anisotropic Material Properties of Abdominal Aortic Aneurysms In Vivo Hegner, Achim Maximilian Claus An abdominal aortic aneurysm (AAA) is a permanent local dilatation of the human abdominal aorta, posing a severe health risk due to a rupture related mortality rate of up to 85 percent. Current clinical risk assessment relies on a population derived threshold value for the maximum aneurysm diameter. However, this global metric ignores local tissue alterations from remodeling processes affecting the biomechanical properties of the vessel wall. These changes create heterogeneous deformations and vulnerable regions prone to rupture. Consequently, the diameter based criterion allows limited individual risk stratification, risking unnecessary surgeries or missed critical cases. Currently, no clinically established patient specific biomarker reliably predicts rupture risk. This work focuses on developing and validating non invasive, in vivo applicable methods to characterize the elastic properties of different AAA tissue regions. Additionally, it pursues material behavior identification using a hyperelastic and anisotropic material model. The goal is to provide patient specific markers alongside the diameter criterion to assess individual rupture risk. The central measurement technique is high resolution 4D ultrasound (3D and time) combined with speckle tracking (4D US), enabling spatially and temporally resolved assessment of local wall deformations. The first study developed a method to improve the reliability of single ultrasound acquisitions, reflecting current clinical standards. A comprehensive analysis approach combined multiple estimates of wall deformation from a single acquisition into an averaged model. Using in vivo data from 10 AAA patients, strains were calculated in wall regions with structurally different properties, such as calcifications. The ability to distinguish these regions using 4D US indicates sufficient resolution to differentiate potentially rupture prone regions from more stable areas. Using the averaged models, all patients could be significantly distinguished based on circumferential strain compared to only 25 percent using individual measurements. This enables reliable individual level statements for the first time. The second study extended the focus to thrombotic regions using data from an animal study involving ten apolipoprotein E deficient mice. By comparing in vivo findings with histological cross sections, a new method assigned measured wall strains to five different tissue regions within the vessel wall and thrombus areas. The analysis demonstrated that 4D US can reliably differentiate between healthy and diseased aortic walls, as well as various tissue types within the wall and thrombus. A third study investigated methods for determining material behavior. An existing inverse in vivo material identification method was fundamentally revised to determine parameters of a hyperelastic and anisotropic material model. The new method is approximately seven times faster and accounts for the thrombus, making it applicable to almost all clinically relevant AAA cases. After in vitro validation on a porcine aorta, in vivo validation compared preoperatively identified material parameters with those from uniaxial tensile tests on ex vivo tissue samples from two AAA patients, showing good agreement. This thesis thus advances the development of patient specific biomarkers to assess acute rupture risk in AAA patients clinically. It strengthens the potential of 4D ultrasound as a non invasive technique for high resolution quantification of structural vessel wall changes. Combining local deformation analysis and material identification opens new perspectives for understanding disease progression and improving clinical decision making and patient outcomes. Thu, 01 Jan 2026 00:00:00 GMT http://hdl.handle.net/10498/39996 2026-01-01T00:00:00Z http://hdl.handle.net/10498/37187 Desarrollo de materiales de base polimérica sostenibles orientado al diseño de productos en fabricación aditiva de gran formato Burgos Pintos, Pedro La creación de nuevos productos está en constante evolución, impulsada por la búsqueda de procesos de fabricación más sostenibles. Para lograr esta mayor sostenibilidad existen numerosas estrategias, que debidamente combinadas y optimizadas, permiten alcanzar este objetivo. Dentro de estas estrategias se encuentran el desarrollo y uso de materiales más sostenibles, empleando residuos de origen natural, materiales reciclados, o materiales biobasados y compostables. Así se logra el ahorro de materias primas poliméricas. Otras estrategias son la fabricación por demanda, sin acumulación de stocks innecesarios, así como la disminución del material desperdiciado en el proceso. Todos estos factores se pueden lograr empleando diferentes tecnologías de Fabricación Aditiva (FA), gracias a su versatilidad. A través de estas tecnologías se pueden lograr geometrías complejas reduciendo costes de utillajes adicionales y desperdicios de materiales, así como una fabricación a medida sin que esto implique un aumento en el coste del producto final, para pequeños lotes de producción. Estas tecnologías se están integrando en todo tipo de sectores industriales, dentro del ámbito de la Industria 4.0. La Fabricación Aditiva de Gran Formato (LFAM), realizada a través de tecnologías de Extrusión de Material, se está haciendo presente en numerosos sectores industriales como la automoción, el aeronáutico, el naval y el del hábitat. Con este tipo de tecnologías se pueden fabricar piezas de volúmenes superiores al metro cúbico, reduciendo costes de utillajes adicionales y con tiempos de fabricación inferiores a otras técnicas. Todo ello con la capacidad de lograr fabricar piezas de alta complejidad geométrica. Además, permite emplear todo tipo de materiales, en función de los requisitos que la pieza necesite. En este contexto, un objetivo de esta tesis doctoral es el estudio del uso de materiales que habitualmente se usan en procesos de fabricación como inyección y extrusión (PET), para aprovechar sus excelentes propiedades mecánicas en piezas fabricadas en LFAM, gracias a su naturaleza semicristalina. En esta tesis también se estudia el desarrollo de materiales de carácter sostenible que incorporan agro-residuos (corcho y hueso de aceituna), con el fin de reducir el uso de materiales plásticos derivados del petróleo (PETG y ASA) y para valorizar residuos en forma de nuevos productos. Además, para poder ofrecer aplicaciones concretas y demostradores de estos materiales en productos concretos, se desarrolla una metodología para el diseño de productos en LFAM. Para ello se emplean softwares de modelado paramétrico, que permiten obtener geometrías complejas, fácilmente variables, que aprovechen la versatilidad de la FA. De esta manera se optimizará el proceso de fabricación para lograr productos funcionales con la menor cantidad de defectos propios de este tipo de fabricación y personalizados en función de su aplicación y estética. De esta forma se han desarrollado materiales compuestos empleando matrices poliméricas derivadas del petróleo y agro residuos del corcho y del hueso de aceituna procesables en LFAM, cuyas propiedades mecánicas les permiten ser empleados en sectores como el hábitat. Además, se ha demostrado la posibilidad de usar el material PET en LFAM, ampliando el rango de materiales aptos para su uso en FA, con excelentes propiedades mecánicas y térmicas. Wed, 09 Apr 2025 00:00:00 GMT http://hdl.handle.net/10498/37187 2025-04-09T00:00:00Z http://hdl.handle.net/10498/29454 Variations on bayesian optimization applied to numerical flow simulations Larroque, Anthony Bayesian Optimization (BO) has recently regained interest in optimization problems involving expensive black-box objective functions. Several variants have been proposed in the literature, such as including gradient and/or multi-fidelity information, and it has been extended to multi-objective optimization problems. Despite its recent applications to numerical flow simulations, the efficiency of this method and its variants remains to be characterized in typical applications involving canonical flows. In this work, the efficiency of classical BO and alternative derivative-free methods is compared on a simplified flow case, i.e. drag reduction in the two-dimensional flow around a cylinder. The application of BO to complex flows is then showcased by considering a three-dimensional case at Reynolds number Re = 3900. Next, the performance of BO with gradient and/or multi-fidelity information is investigated for global modelling and optimization on typical benchmark objective functions and on the cylinder case at Re = 200. Finally, an algorithm combining dimension reduction and Multi-objective Bayesian Optimization (MOBO) is proposed. It is found that BO was more efficient than other derivative-free alternatives and showed promising results on the three-dimensional cylinder at Re = 3900 by reducing drag by 23 %. The performance of the algorithm was further improved when multi-fidelity and/or gradient information was included, both for modelling and optimization. Including gradient information on the low-fidelity model was useful for global modelling and to decrease rapidly the objective function in a BO framework. On the contrary, adding derivative information on the high-fidelity model generally gave the most accurate approximation of the minimum but was inefficient for global modelling when the computational cost of the gradient was high. Finally, the developed algorithm combining dimension reduction and MOBO enabled us to obtain more precise and diverse minima. Sun, 01 Jan 2023 00:00:00 GMT http://hdl.handle.net/10498/29454 2023-01-01T00:00:00Z