Electronic Thesis and Dissertation Repository

Degree

Doctor of Philosophy

Program

Biomedical Engineering

Supervisor

Dr. Abbas Samani

Abstract

To model cardiac mechanics effectively, various mechanical characteristics of cardiac muscle tissue including anisotropy, hyperelasticity, and tissue active contraction characteristics must be considered. Some of these features cannot be implemented using commercial finite element (FE) solvers unless additional custom-developed computer codes/subroutines are appended. Such codes/subroutines are unavailable for the research community. Accordingly, the overarching objective of this research is to develop a novel LV mechanics model which is implementable in commercial FE solvers and can be used effectively within inverse FE frameworks towards cardiac disease diagnosis and therapy. This was broken down into a number of objectives. The first objective is to develop a novel cardiac tissue mechanical model. This model was constructed of microstructural cardiac tissue constituents while their associated volume contributions and mechanical properties were incorporated into the model. These constituents were organized in small FE tissue specimen models consistent with the normal/pathological cardiac tissue microstructure. In silico biaxial/uniaxial mechanical tests were conducted on the specimen models and corresponding stress-strain data were validated by comparing them with cardiac tissue data reported in the literature. Another objective of this research is developing a novel FE-based mechanical model of the LV which is fully implementable using commercial FE solvers without requiring further coding, potentially leading to a computationally efficient model which is easily adaptable to diverse pathological conditions. This was achieved through considering a novel composite material model of the cardiac tissue while all aspects of the cardiac mechanics including hyperelasticity, anisotropy, and active tissue responses were preserved. The model was applied to an in silico geometry of a canine LV under both normal and pathological conditions and systolic/diastolic responses of the model were compared with corresponding data of other LV mechanical models and LV contraction measurements. To test the suitability of the proposed cardiac model for FE inversion-based algorithms, the model was utilized for LV diastolic mechanical simulation to estimate the tissue stiffness and blood pressure using an ad-hoc optimization scheme. This led to reasonable tissue stiffness and blood pressure values falling within the range of LV measurements of healthy subjects, confirming the efficacy of this model for inversion-based diagnosis applications.

Share

COinS