Bozzi Silvia, Morbiducci Umberto, Gallo Diego, Ponzini Raffaele, Rizzo Giovanna, Bignardi Cristina, Passoni Giuseppe
a Department of Electronics, Information Science, and Bioengineering , Politecnico di Milano , Milan , Italy.
b Department of Mechanical and Aerospace Engineering , Politecnico di Torino , Turin , Italy.
Comput Methods Biomech Biomed Engin. 2017 Aug;20(10):1104-1112. doi: 10.1080/10255842.2017.1334770. Epub 2017 May 29.
This study investigates the impact that uncertainty in phase contrast-MRI derived inlet boundary conditions has on patient-specific computational hemodynamics models of the healthy human thoracic aorta. By means of Monte Carlo simulations, we provide advice on where, when and how, it is important to account for this source of uncertainty. The study shows that the uncertainty propagates not only to the intravascular flow, but also to the shear stress distribution at the vessel wall. More specifically, the results show an increase in the uncertainty of the predicted output variables, with respect to the input uncertainty, more marked for blood pressure and wall shear stress. The methodological approach proposed here can be easily extended to study uncertainty propagation in both healthy and pathological computational hemodynamic models.
本研究调查了相位对比磁共振成像得出的入口边界条件中的不确定性对健康人体胸主动脉的患者特异性计算血流动力学模型的影响。通过蒙特卡洛模拟，我们就何处、何时以及如何考虑这种不确定性来源提供了建议。研究表明，这种不确定性不仅会传播到血管内流动，还会传播到血管壁处的剪应力分布。更具体地说，结果表明，相对于输入不确定性，预测输出变量的不确定性有所增加，血压和壁面剪应力的情况更为明显。这里提出的方法可以很容易地扩展到研究健康和病理计算血流动力学模型中的不确定性传播。
