Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, OH, United States of America.
Computational Biomodeling (CoBi) Core, Lerner Research Institute, Cleveland Clinic, Cleveland, OH, United States of America.
PLoS One. 2023 Apr 21;18(4):e0284721. doi: 10.1371/journal.pone.0284721. eCollection 2023.
Emergent mechanics of musculoskeletal extremities (surface indentation stiffness and tissue deformation characteristics) depend on the underlying composition and mechanics of each soft tissue layer (i.e. skin, fat, and muscle). Limited experimental studies have been performed to explore the layer specific relationships that contribute to the surface indentation response. The goal of this study was to examine through statistical modeling how the soft tissue architecture contributed to the aggregate mechanical surface response across 8 different sites of the upper and lower extremities. A publicly available dataset was used to examine the relationship of soft tissue thickness (fat and muscle) to bulk tissue surface compliance. Models required only initial tissue layer thicknesses, making them usable in the future with only a static ultrasound image. Two physics inspired models (series of linear springs), which allowed reduced statistical representations (combined locations and location specific), were explored to determine the best predictability of surface compliance and later individual layer deformations. When considering the predictability of the experimental surface compliance, the physics inspired combined locations model showed an improvement over the location specific model (percent difference of 25.4 +/- 27.9% and 29.7 +/- 31.8% for the combined locations and location specific models, respectively). While the statistical models presented in this study show that tissue compliance relies on the individual layer thicknesses, it is clear that there are other variables that need to be accounted for to improve the model. In addition, the individual layer deformations of fat and muscle tissues can be predicted reasonably well with the physics inspired models, however additional parameters may improve the robustness of the model outcomes, specifically in regard to capturing subject specificity.
肌肉骨骼四肢的突发力学（表面凹陷硬度和组织变形特性）取决于各软组织层（即皮肤、脂肪和肌肉）的基础组成和力学特性。已经进行了有限的实验研究来探索有助于表面凹陷反应的特定于层的关系。本研究的目的是通过统计建模来检查软组织结构如何促成上肢和下肢 8 个不同部位的总体机械表面反应。使用公开可用的数据集来检查软组织厚度（脂肪和肌肉）与大块组织表面顺应性的关系。模型仅需要初始组织层厚度，因此将来仅使用静态超声图像即可使用它们。探索了两种受物理启发的模型（一系列线性弹簧），这些模型允许进行简化的统计表示（组合位置和位置特定），以确定表面顺应性和后来的各个层变形的最佳可预测性。在考虑实验表面顺应性的可预测性时，受物理启发的组合位置模型显示出比位置特定模型更好的可预测性（组合位置和位置特定模型的实验表面顺应性的百分比差异分别为 25.4 +/- 27.9%和 29.7 +/- 31.8%）。尽管本研究中提出的统计模型表明组织顺应性取决于各层厚度，但显然还需要考虑其他变量以改进模型。此外，脂肪和肌肉组织的各个层变形可以用受物理启发的模型很好地预测，但是附加参数可能会提高模型结果的稳健性，特别是在捕获个体特异性方面。
