Researchers have developed a new patient-specific Bayesian inverse modeling framework to estimate the mechanical properties of abdominal aortic aneurysm (AAA) walls and the associated intraluminal thrombus (ILT). Recently published in the Annals of Biomedical Engineering, the study utilizes in vivo magnetic resonance imaging (MRI) data to provide a patient-specific assessment of aneurysm mechanics.
To address this, the research team constructed finite element models (FEMs) for four distinct AAA patients using vascular geometry directly derived from MRI scans. By integrating Bayesian optimization with Gaussian Process Regression, they successfully identified complex nonlinear anisotropic material parameters that could accurately replicate the intricate biomechanical behavior of the diseased vessels in a controlled computational environment.
To validate the proposed framework, the team performed rigorous equibiaxial mechanical tests on tissue samples excised from the same patients. The Bayesian-identified parameters demonstrated a close match to the experimental data, successfully reproducing both the measured magnitudes of displacement and the observed anisotropy of the aortic tissues.
The study recorded low root mean square error (RMSE) values when comparing the finite element model-predicted displacements to those directly derived from MRI measurements. The researchers highlighted these findings as evidence of high model fidelity.
The investigation also provided new insights into the mechanical role of the intraluminal thrombus. Peak wall stress distributions revealed that the presence of the ILT significantly attenuates wall stress, with higher concentrations of stress typically localized in those regions of the aorta free of the thrombus. This confirms the functional impact the thrombus has on the overall mechanical state of the aneurysm.
Sensitivity analyses were performed to evaluate how specific variations in wall thickness and the presence of the ILT influence predicted wall stresses. While moderate variations in wall thickness and material properties had a limited impact on the precise locations of peak stress, the total exclusion of the ILT from the computational models resulted in a marked, significant increase in the predicted wall stresses.
The researchers conclude that this Bayesian framework offers a robust, individualized methodology for accurately estimating material properties. By providing reliable predictions of wall stress distributions and emphasizing the protective role of the ILT, this approach could potentially assist clinicians in improving the assessment of rupture risk for patients managing abdominal aortic aneurysms, moving closer to truly individualized care.
Subject of Research: Computational biomechanical assessment of abdominal aortic aneurysm wall and thrombus properties.
Article Title: An MRI-Based Framework for Patient-Specific Abdominal Aortic Aneurysm Mechanics Assessment
Article References: Caruso, G., Dux-Santoy, L., Guala, A., Bragulat-Arévalo, M., Teixidó-Turà, G., García-Reyes, M., Rodriguez-Palomares, J. F., Bellmunt, S., Martínez, M. Á., & Peña, E. (2026). An MRI-Based Framework for Patient-Specific Abdominal Aortic Aneurysm Mechanics Assessment. Annals of Biomedical Engineering. https://doi.org/10.1007/s10439-026-04459-1
Image Credits: AI Generated
DOI: 10.1007/s10439-026-04459-1
Keywords: Abdominal Aortic Aneurysm, AAA, Biomechanics, MRI, Bayesian Inference, Gaussian Process Regression, Finite Element Analysis, Medical Imaging, Arterial Mechanics, Intraluminal Thrombus, Vascular Health, Patient-Specific Modeling, Computational Modeling, Mechanical Properties, Aneurysm Rupture Risk, Medical Physics, Biomedical Engineering
News Source: Denise Maddox. (November 18, 2026). Bayesian Framework Developed and Validated for Patient-Specific Abdominal Aortic Aneurysm Mechanics. Scienmag.



