Introduction

There is a striking difference in the reported mean response of abdominal aortic aneurysm tissue in academic literature depending on the type of tests (uniaxial vs biaxial) performed. In this paper, the hypothesis variability caused by differences in experimental protocols is explored using porcine aortic tissue as a substitute for aneurysmal tissue.

Methods

Nine samples of porcine aorta were created and both uniaxial and biaxial tests were performed. Three effects were investigated. (i) Effect of sample (non) preconditioning, (ii) effect of objective function used (normalised vs non-normalised), and (iii) effect of chosen procedure used for mean response calculation: constant averaging (CA) vs fit to averaged response (FAR) vs fit to all data (FAD). Both the overall shape of mean curve and mean initial stiffness were compared.

Results

(i) Non-preconditioning led to a much stiffer response, and initial stiffness was about three times higher for a non-preconditioned response based on uniaxial data compared to a preconditioned biaxial response. (ii) CA led to a much stiffer response compared to FAR and FAD procedures which gave similar results. (iii) Normalised objective function produced a mean response with six times lower initial stiffness and more pronounced nonlinearity compared to non-normalised objective function.

Discussion

It is possible to reproduce a mechanically inconsistent response purely by using the chosen experimental protocol. Non-preconditioned data from failure tests should be used for FE simulation of the elastic response of aneurysms. CA should not be used to obtain a mean response.

中文翻译：

---

腹主动脉瘤组织平均机械反应不一致的原因

介绍

在学术文献中，所报告的腹部主动脉瘤组织的平均反应差异很大，这取决于所进行的测试类型（单轴与双轴）。在本文中，使用猪主动脉组织替代动脉瘤组织来探索由实验方案差异引起的假设变异性。

方法

制作了九个猪主动脉样品，并进行了单轴和双轴测试。研究了三种效果。（i）样本（非预处理）的影响，（ii）使用的目标函数的影响（归一化与非归一化），以及（iii）用于平均响应计算的所选过程的影响：恒定平均（CA）vs拟合平均值响应（FAR）vs适合所有数据（FAD）。比较平均曲线的整体形状和平均初始刚度。

结果

（i）非预处理导致刚性更大，基于单轴数据的非预处理响应的初始刚度比预处理双轴响应高约三倍。（ii）与给出类似结果的FAR和FAD程序相比，CA导致的响应更加僵硬。（iii）与未归一化的目标函数相比，归一化的目标函数产生的平均响应具有低六倍的初始刚度和更明显的非线性。

讨论区

完全可以通过使用所选的实验方案来重现机械不一致的响应。来自故障测试的非预处理数据应用于有限元模拟动脉瘤的弹性反应。CA不应用于获得平均响应。