Med Biol Eng Comput. 2025 Oct 7. doi: 10.1007/s11517-025-03459-y. Online ahead of print.
ABSTRACT
Over the past few years, there has been an increase of clinical interest aimed at looking for correlations between morphology, extracted through statistical shape models (SSMs), and hemodynamics, extracted through computational fluid dynamics (CFD) simulations, in cardiovascular diseases. This study explores correlations between aortic morphology and hemodynamics in the thoracic aorta (TA). Existing research often simplifies geometries by excluding supra-aortic vessels due to software limitations in non-rigid registration. To overcome this, a novel algorithm was used to include these vessels in TA analysis. Principal component analysis reduced dimensionality, followed by automatic CFD simulations and correlation analysis between geometric and hemodynamic parameters. The first ( M 0 ) and second ( M 1 ) SSM modes explained 46.9 % and 22.4 % of dataset variance, respectively. Significant correlations were identified between M 0 and ascending TA aneurysm volume (Pr = 0.69), and M 1 and TA tortuosity (Pr = 0.60). Ten TA shapes were generated by varying standard deviations of M 0 and M 1 from -2 to +2, and CFD simulations revealed links between wall shear stress (WSS) indicators and TA morphology. This study presents a novel pipeline to analyze geometric and hemodynamic correlations using realistic TA geometries generated via SSM.
PMID:41055863 | DOI:10.1007/s11517-025-03459-y
