Project Description
A patient-specific, fluid–structure interaction framework for the functional assessment of ascending thoracic aortic aneurysms — informed by clinical imaging, ex-vivo mechanics and uncertainty quantification.
Abstract
Thoracic aortic aneurysm (TAA) is a disease with an incidence of about 5-10/100,000 persons per year. Its treatment has been evolving over the last decades. Advances in surgical techniques and functional imaging have paved the way of modern clinical medicine with a significant social and economic impact. The European Societies of Cardiology and Cardiothoracic Surgery advocate, for aneurysms with diameter below 55 mm, the surgical intervention in the presence of certain risk factors: coarctation of the aorta, arterial hypertension, family history for aortic dissection and rapid growth (> 3 mm / year).
Despite these European guidelines, fundamental clinical open questions do persist because the disease prevention and diagnosis are not completely effective and reliable. There are controversies regarding the pathogenesis and the true risk of aneurysm rupture since more than half of the patients, in whom aortic dissection occurred, had an aortic diameter considered normal (between 40 mm and 50 mm) at the time of the event, and therefore with no indication of surgery.
It has been accepted that the formation of the aneurysm has a multifactorial etiology. Nonetheless, there are two factors considered to be predominant: genetic and hemodynamic. Observations have confirmed that this disease is accelerated by changes in hemodynamic, yielding high shear stresses in the aortic wall. Those values promote the degradation of elastin fibres and, consequently, the necrosis of the middle layer of the aortic wall, which results in the progressive loss of the aorta elasticity and its dilation over time, eventually leading to aortic dissection.
In addition, the shear stresses in the arterial wall, that can lead to premature rupture, are dependent not only on the geometry of the aneurysm, but also on the structure of the aortic tissue, the state of mechanical properties of the wall and the hemodynamic boundary conditions.
From Bedside to In-Silico
The project integrates clinical, computational and experimental standpoints into a single pipeline that turns patient data into a personalised risk indicator.
Research Hypothesis
The hypothesis formulated in this project is that TAA treatment can benefit from a personalised patient-specific computational modelling framework, based on fluid–structure interaction (FSI) and informed by clinical, imaging, and ex-vivo deformation data.
Experimental Approach
Ex-vivo experiments and histopathology observations over TAA tissue are proposed from donor patients who have undergone surgery, with formal recommendation and following ethical agreements. New experimental protocols will be proposed to test the aorta tissue. The material deformation will be accessed by an advanced image-based monitoring technique. Inverse identification strategies, based on equilibrium equations and optimization between numerical and experimental mechanical responses, are proposed to extract relevant constitutive properties.
The identification of local properties for the constitutive model can then be statistically correlated with histopathological and radiological observations to establish relationships between structure and mechanical properties. These results will be used to develop mathematical models that will be implemented in the hemodynamic simulation tool for the functional assessment of TAA.
Computational Framework
From a clinical point of view, although the FSI models have the advantage to be oriented towards patient-specific medicine, a major criticism and limitation is their deterministic structure. This is established in terms of geometrical and mechanical input parameters of the digital twin that do not account effectively for uncertainties on the physical biological system.
To overcome this drawback, model updating methods for uncertainty quantification and validation will be introduced and predictive responses analysed from a statistical standpoint. The model unrealistic predictive indications and errors will be analysed from a stochastic inverse problem perspective in view of reducing the epistemic sources of uncertainty.
Key Objectives
Develop patient-specific computational models for ascending thoracic aortic aneurysms.
Characterize mechanical properties of aortic tissue through ex-vivo experiments.
Implement fluid–structure interaction simulations.
Create uncertainty quantification frameworks for clinical decision support.
Establish correlations between biomechanical markers and clinical outcomes.