A Combination of Constitutive Damage Model and Artificial Neural Networks to Characterize the Mechanical Properties of the Healthy and Atherosclerotic Human Coronary Arteries. Issue 9 (2nd February 2017)

• Record Type: Journal Article
• Authors: Karimi, Alireza; Rahmati, Seyed Mohammadali; Sera, Toshihiro; Kudo, Susumu; Navidbakhsh, Mahdi
• Is Part Of: Artificial organs. Volume 41:Issue 9(2017)
• Page Start: E103
• Page End: E117
• Publication Date: 2017-02-02
• Subjects: Coronary artery -- —Atherosclerosis -- —Constitutive damage model -- —Collagen fiber remodeling -- —Histological structure; Artificial organs -- Periodicals; 617.956
• DOI: 10.1111/aor.12855
• Languages: English
• ISSNs: 0160-564X

Abstract

Abstract: It has been indicated that the content and structure of the elastin and collagen of the arterial wall can subject to a significant alteration due to the atherosclerosis. Consequently, a high tissue stiffness, stress, and even damage/rupture are triggered in the arterial wall. Although many studies so far have been conducted to quantify the mechanical properties of the coronary arteries, none of them consider the role of collagen damage of the healthy and atherosclerotic human coronary arterial walls. Recently, a fiber family‐based constitutive equation was proposed to capture the anisotropic mechanical response of the healthy and atherosclerotic human coronary arteries via both the histostructural and uniaxial data. In this study, experimental mechanical measurements along with histological data of the healthy and atherosclerotic arterial walls were employed to determine the constitutive damage parameters and remodeling of the collagen fibers. To do this, the preconditioned arterial tissues were excised from human cadavers within 5‐h postmortem, and the mean angle of their collagen fibers was precisely determined. Thereafter, a group of quasistatic axial and circumferential loadings were applied to the arterial walls, and the constrained nonlinear minimization method was employed to identify the arterial parameters according to the axial and circumferential extension data. The remodeling of the collagen fibers during the tensile test was also predicted via Artificial Neural Networks algorithm. Regardless of loading direction, the results presented a noteworthy load‐bearing capability and stiffness of the atherosclerotic arteries compared to the healthy ones ( P < 0.005). Theoretical fiber angles were found to be consistent with the experimental histological data with less than 2 and 5° difference for the healthy and atherosclerotic arterial walls, respectively. The pseudoelastic damage model data were also compared with that of the experimental data, and interestingly, the arterial mechanical behavior for both the primary loading (up to the elastic region) and the discontinuous softening (up to the ultimate stress) was well addressed. The proposed model predicted well the mechanical response of the arterial tissue considering the damage of collagen fibers for both the healthy and atherosclerotic arterial walls.