Uncertainty quantification in subject-specific estimation of local vessel mechanical properties

Area of Science:

• Biomechanics
• Biomedical Engineering
• Cardiovascular Research

Background:

• Accurate quantification of local mechanical properties is crucial for understanding blood vessel homeostasis.
• Panoramic digital image correlation (pDIC) enables high-fidelity 3D reconstructions of small animal vessels.
• Previous inverse modeling lacked uncertainty quantification for subject-specific biomechanical property estimation.

Purpose of the Study:

• To integrate a novel uncertainty quantification and propagation pipeline into existing inverse modeling for pDIC data.
• To enable systematic reporting of uncertainties in local mechanical property estimates of blood vessels.
• To facilitate subject-specific and group-level statistical analyses of vessel wall mechanics.

Main Methods:

• Integrated empirical and analytic Bayesian techniques for uncertainty quantification and propagation.
• Applied the enhanced inverse modeling workflow to ascending thoracic aorta data from three mouse models.
• Quantified uncertainties in constitutive model parameters and circumferential/axial tangent stiffness.

Main Results:

• Successfully quantified uncertainties in local mechanical properties of the mouse aorta.
• Demonstrated the ability to systematically report parameter uncertainties.
• Enabled robust subject-specific and group-level statistical comparisons of vessel mechanics.

Conclusions:

• The developed pipeline enhances the reliability of biomechanical property estimation from pDIC data.
• This approach provides crucial uncertainty metrics for subject-specific and comparative analyses of vascular mechanics.
• The methodology advances our understanding of blood vessel mechanical homeostasis and disease.