Multi-patient study for coronary vulnerable plaque model comparisons: 2D/3D and fluid-structure interaction...

Area of Science:

• Biomedical Engineering
• Computational Mechanics
• Cardiovascular Research

Background:

• Image-based computational models are crucial for analyzing atherosclerotic plaque progression and vulnerability.
• Quantifying prediction differences across multi-patient computational models remains a challenge.

Purpose of the Study:

• To compare various computational models for plaque analysis.
• To quantify the impact of 2D simplification, circumferential shrink, fluid-structure interactions (FSI), and cyclic bending on plaque wall stress/strain (PWS/PWSn) and flow shear stress (FSS).

Main Methods:

• Acquired in vivo intravascular ultrasound (IVUS) coronary plaque data from seven patients.
• Constructed seven 2D/3D models with and without specific factors (shrink, bending, FSI) for each patient.
• Compared PWS/PWSn and FSS calculations across models using patient data (388 slices).

Main Results:

• 2D models without shrink overestimated PWS by 17.26% compared to those with shrink.
• 3D FSI models with cyclic bending showed PWS changes of 15.07%-49.52% (avg. 30.13%) at high curvature areas.
• Flow-only models predicted higher FSS (4.02%-11.29%) than full FSI models; PWS/PWSn differences between FSI and structure-only models were minimal (4.38%, 1.78%).

Conclusions:

• Model differences in plaque analysis exhibit patient-specific variations.
• Structure-only models provide accurate approximations for PWS/PWSn, offering computational efficiency.
• FSI and flow-only model differences are more pronounced for minimum FSS, relevant to plaque progression.