Abstract

This project continues to fit the experimental data on the stress-strain behavior of newborn mice’s ascending aorta. The stress-strain behavior of newborn mouse aorta is highly nonlinear because of its unique composition of elastin and collagen in the arterial wall. To mathematically describe this behavior, more strain energy functions of different biological materials were used in fitting. Multiple error functions were used to improve accuracy. K-fold cross-validation was applied in regression to avoid overfitting. In addition to the data of aorta with elastin knockout, those of aorta with lysyl oxidase knockout and fibulin 4 knockout are also evaluated in this project using the strain energy functions found. It is noteworthy that in this report, acronyms for these types of aorta are used. Elastin knockout = ELN KO, Fibulin 4 Knockout = FIB 4 KO, Lysyl Oxidase knockout = LOX KO, Wild Type = WT. Regression results from MATLAB shows that while the strain energy functions provide accurate fitting overall, improvements on the regression model can still be made to minimize the errors in cross validation. The strain energy and stiffness in the loading cycle are computed using the results returned by MATLAB: the strain energy for aorta with genetic defects is smaller than that for wild type samples. Comparing this difference may help to understand cardiovascular disease related to genetic defects.

Document Type

Final Report

Author's School

McKelvey School of Engineering

Author's Department

Mechanical Engineering and Materials Science

Class Name

Mechanical Engineering and Material Sciences Independent Study

Date of Submission

12-21-2021

Share

COinS