Osteoarthritis is a whole joint disease with cartilage degeneration being an important manifestation. Tissue engineering treatment is a solution for repairing cartilage defects by implantation of chondrocyte-laden hydrogel constructs within the defect. In silico models have recently been introduced to simulate and optimize the design of these constructs. These models require accurate knowledge on the mechanical properties of the hydrogel constructs and cartilage explants, which are challenging to obtain due to their anisotropic structure and time-dependent behaviour.

We performed a systematic in silico parameter sensitivity analysis to find the most efficient unconfined compression testing protocols for mechanical characterization of hydrogel constructs and cartilage explants, with a minimum number of tests but maximum identifiability of the material parameters. The construct and explant were thereby modelled as porohyperelastic and fibril-reinforced poroelastic materials, respectively. Three commonly used loading regimes were simulated in Abaqus (ramp, relaxation and dynamic loading) with varying compressive strain magnitudes and rates. From these virtual experiments, the resulting material parameters were obtained for each combination using a numerical inverse identification scheme.

For hydrogels, maximum sensitivity to the different material parameters was found when using a single step ramp loading (20% compression with 10%/s rate) followed by 15 min relaxation. For cartilage explants, a two-stepped ramp loading (10% compression with 10%/s rate and 10% compression with 1%/s rate), each step followed by 15 min relaxation, yielded the maximum sensitivity to the different material parameters. With these protocols, the material parameters could be retrieved with the lowest amount of uncertainty (hydrogel: < 2% and cartilage: < 6%). These specific results and the overall methodology can be used to optimize mechanical testing protocols to yield reliable material parameters for in silico models of cartilage and hydrogel constructs.

骨关节炎是一种以软骨退变为重要表现的全关节疾病。组织工程治疗是通过在缺损内植入富含软骨细胞的水凝胶结构来修复软骨缺损的解决方案。最近引入了计算机模拟模型来模拟和优化这些结构的设计。这些模型需要对水凝胶结构和软骨外植体的机械特性有准确的了解，由于它们的各向异性结构和时间依赖性行为，很难获得这些知识。

我们进行了系统的计算机参数敏感性分析，以找到最有效的无侧限压缩测试方案，用于水凝胶结构和软骨外植体的机械表征，测试次数最少，但材料参数的可识别性最大。因此，构建体和外植体分别建模为多孔超弹性和原纤维增强的多孔弹性材料。在 Abaqus 中模拟了三种常用的加载方式（斜坡、松弛和动态加载），具有不同的压缩应变大小和速率。从这些虚拟实验中，使用数值逆识别方案获得每种组合的所得材料参数。

对于水凝胶，当使用单步斜坡加载（20% 压缩，10%/s 速率）然后松弛 15 分钟时，发现对不同材料参数的最大敏感性。对于软骨外植体，两步斜坡加载（10% 压缩以 10%/s 速率和 10% 压缩以 1%/s 速率），每一步后松弛 15 分钟，产生对不同材料参数的最大敏感性。使用这些协议，可以以最低的不确定性（水凝胶：< 2% 和软骨：< 6%）检索材料参数。这些特定的结果和整体方法可用于优化机械测试协议，为软骨和水凝胶结构的silico模型产生可靠的材料参数。