The aorta is characterized by a remarkable hemodynamic richness in its thoracic segment, mainly dictated by proximity to the beating heart, aortic wall motion, and complex anatomy (Jin et al., 2003, Sengupta et al., 2008, Yearwood and Chandran, 1982). Such hemodynamic richness has contributed to make the aorta a site of election for fluid mechanics studies aimed at highlighting the links of blood transport with cardiovascular pathophysiology (Garcia et al., 2018, Guala et al., 2022, Markl et al., 2004, Morbiducci et al., 2011, Stokes et al., 2021, Vignali et al., 2021). At the same time, aortic hemodynamics is characterized by a large spatiotemporal heterogeneity that has contributed to hamper a robust, univocal definition of the large-scale coherent fluid structures characterizing normal aortic blood flow, and able to discriminate physiology from pathology (Calò et al., 2023).

In the last decades, the coupling of medical imaging and computational fluid dynamics (CFD) has gained momentum as an effective tool for studying the aortic hemodynamics with high spatiotemporal resolution in personalized in silico models (Antonuccio et al., 2021, Cilla et al., 2020, De Nisco et al., 2020, Gallo et al., 2012, Morbiducci et al., 2013, Romarowski et al., 2018). However, CFD models require assumptions which introduce important sources of uncertainty and might entail the generality of CFD-based results as well as their translation to clinics. In personalized CFD aortic models, a main source of uncertainty is represented by assumptions on wall displacements. In this regard, a largely adopted model idealization is the rigid-wall assumption, even if some studies suggest that incorporating MRI-measured subject-specific wall displacements into CFD simulations provides better agreement in terms of velocity fields with in vivo observations (Jin et al., 2003, Lantz et al., 2014). Two approaches alternative to rigid-wall CFD simulations can be implemented to account for aortic wall displacements. The first one is based on the classical fluid–structure interaction (FSI) modelling strategy (Caballero and Laín, 2015, Mirramezani and Shadden, 2022). However, FSI requires additional assumptions on wall material properties, which cannot be easily determined in vivo and therefore represent further sources of uncertainty. Moreover, FSI is characterized by an increased complexity and computational efforts hindering its clinical translation. To overcome FSI inherent limitations, an alternative approach based on the moving-boundary method (MBM) has emerged. MBM relies on the integration of clinical image-based measurements of wall displacements along the cardiac cycle and CFD simulations in the same computational framework (Capellini et al., 2020, Lantz et al., 2019, Torii et al., 2010), allowing to incorporate subject-specific wall displacements in simulations at lower computational costs and without posing any hypothesis on vessels wall material properties. Among the MBM strategies applied to in silico aortic models (Bonfanti et al., 2018, Jin et al., 2003, Lantz et al., 2014, Stokes et al., 2021), a framework integrating dynamic CT images, transient CFD simulations and a mesh morphing technique based on Radial Basis Functions (RBF) was recently introduced (Capellini et al., 2020). The RBF-based mesh morphing strategy has the advantage of keeping mesh connectivity avoiding re-meshing at each time-step of the simulation, which would result in a considerable increase in computational time. Moreover, the RBF-based morphing strategy has proven its capability in reproducing large deformations like those characterizing ascending thoracic aortic aneurysm progression, while preserving mesh quality (Biancolini et al., 2020, Capellini et al., 2018).

The overarching hypothesis of this study is that wall displacements might impact the coherence of large-scale blood flow structures in the ascending aorta (AAo) in such a way that might not be accounted for assuming rigid walls in aortic CFD simulations. Interpreting the large-scale AAo hemodynamics in terms of flow coherence finds its rationale in the effectiveness of a coherence-based analysis in characterizing physiologically relevant conditions such as flow recirculation, separation and reattachment, local heterogeneity in the flow field, and mechanisms governing stirring and mixing (Shadden and Taylor, 2008). The above-mentioned hypothesis is tested on three subject-specific models of healthy human thoracic aorta performing two simulations on each reconstructed anatomy, assuming: (1) aortic rigid walls, and (2) subject-specific aortic wall displacements imposed applying an RBF-based mesh morphing strategy (Capellini et al., 2020). The influence of wall displacements on the AAo large-scale fluid structures is analyzed in terms of through-plane velocity coherence and in-plane velocity distribution, as well as in terms of helical flow because of its recognized physiological significance in cardiovascular flows (Liu et al., 2009, Morbiducci et al., 2011). For the sake of completeness, the study is complemented by the analysis of the impact that aortic wall displacements have on wall shear stress (WSS), given its established role in the initiation and progression of vascular disease.