Department of Mechanical Engineering, University of Rochester, Rochester, NY 14627, USA.
Department of Electrical and Computer Engineering, University of Rochester, Rochester, NY 14627, USA.
Acta Biomater. 2022 Jul 1;146:259-273. doi: 10.1016/j.actbio.2022.04.047. Epub 2022 May 5.
Elastography researchers have utilized several rheological models to characterize soft tissue viscoelasticity over the past thirty years. Due to the frequency-dependent behavior of viscoelastic parameters as well as the different techniques and frequencies employed in various studies of soft tissues, rheological models have value in standardizing disparate techniques via explicit mathematical representations. However, the important question remains: which of the several available models should be considered for widespread adoption within a theoretical framework? We address this by evaluating the performance of three well established rheological models to characterize ex vivo bovine liver tissues: the Kelvin-Voigt (KV) model as a 2-parameter model, and the standard linear solid (SLS) and Kelvin-Voigt fractional derivative (KVFD) models as 3-parameter models. The assessments were based on the analysis of time domain behavior (using stress relaxation tests) and frequency domain behavior (by measuring shear wave speed (SWS) dispersion). SWS was measured over a wide range of frequency from 1 Hz to 1 kHz using three different tests: (i) harmonic shear tests using a rheometer, (ii) reverberant shear wave (RSW) ultrasound elastography scans, and (iii) RSW optical coherence elastography scans, with each test targeting a distinct frequency range. Our results demonstrated that the KVFD model produces the only mutually consistent rendering of time and frequency domain data for liver. Furthermore, it reduces to a 2-parameter model for liver (correspondingly to a 2-parameter "spring-pot" or power-law model for SWS dispersion) and provides the most accurate predictions of the material viscoelastic behavior in time (>98% accuracy) and frequency (>96% accuracy) domains. STATEMENT OF SIGNIFICANCE: Rheological models are applied in quantifying tissues viscoelastic properties. This study is unique in presenting comprehensive assessments of rheological models.
弹性成像研究人员在过去三十年中利用了几种流变学模型来描述软组织的粘弹性。由于粘弹性参数的频率依赖性以及不同研究中使用的技术和频率不同,流变学模型通过显式数学表示在标准化不同技术方面具有价值。然而,重要的问题仍然存在:在理论框架内,应该考虑几种可用模型中的哪一种进行广泛采用?我们通过评估三种成熟的流变学模型来描述牛肝组织的性能来解决这个问题:凯氏双组分模型(KV)作为 2 参数模型,以及标准线性固体(SLS)和凯氏分数阶导数(KVFD)模型作为 3 参数模型。评估基于时域行为(使用应力松弛测试)和频域行为(通过测量剪切波速度(SWS)的色散)的分析。使用三种不同的测试方法在很宽的频率范围内(从 1 Hz 到 1 kHz)测量 SWS:(i)使用流变仪进行谐波剪切测试,(ii)反射剪切波(RSW)超声弹性成像扫描,以及(iii)RSW 光学相干弹性成像扫描,每个测试针对不同的频率范围。我们的结果表明,KVFD 模型为肝产生了唯一的时间和频域数据的相互一致表示。此外,它简化为肝的 2 参数模型(相应地,SWS 色散的 2 参数“弹簧罐”或幂律模型),并提供了最准确的材料粘弹性行为的时间预测(>98%的准确性)和频率预测(>96%的准确性)。意义声明:流变学模型用于量化组织的粘弹性特性。本研究的独特之处在于对流变学模型进行了全面评估。