A Protocol for Coupling Volumetrically Dynamic In-Vitro Experiments to Numerical Physiology Simulation for a Hybrid Cardiovascular Model
A Protocol for Coupling Volumetrically Dynamic In-Vitro Experiments to Numerical Physiology Simulation for a Hybrid Cardiovascular Model
复制标题
混合心血管模型的体积动态体外实验与数值生理模拟耦合方案
DOI：
10.1109/tbme.2022.3216542
复制
发表时间：
2022-10
影响因子：
4.6
通讯作者：
Abraham Umo;Ethan O. Kung
中科院分区：
文献类型：
--
作者：
Abraham Umo;Ethan O. Kung
目标：生理学模拟耦合实验 (PSCOPE) 是一种混合建模框架，使物理流体实验能够在心血管生理学闭环计算模拟的背景下进行。以前的 PSCOPE 方法将刚性实验与生理学的集总参数网络 (LPN) 结合起来，但与流体体积周期性变化的体积动态实验不兼容。我们通过引入一种能够将多分支和体积动态体外实验与 LPN 耦合的方法来解决这一限制。方法：我们提出的方法利用迭代加权平均算法来识别给定 PSCOPE 模型的唯一解波形。我们通过将体外实验的数学替代项直接集成到 LPN 中以得出参考解决方案来确认这些 PSCOPE 解决方案的准确性，该参考解决方案作为验证使用我们提出的方法将相同的数学替代项与 LPN 耦合所获得的解决方案的黄金标准。最后，我们通过将体外肾循环实验与 LPN 结合来说明 PSCOPE 方法的实际应用。结果：与参考溶液相比，流量和压力波形的归一化均方根误差为0.001%∼0.55%，证明了耦合方法的准确性。结论：我们成功地将体外实验与 LPN 结合起来，在物理实验中的传感器和驱动限制的约束下展示了真实世界的性能。意义：本研究引入了一种 PSCOPE 方法，可用于研究表现出周期性体积变化的医疗设备和解剖结构，从而扩展了混合框架的实用性。
Objective: The Physiology Simulation Coupled Experiment (PSCOPE) is a hybrid modeling framework that enables a physical fluid experiment to operate in the context of a closed-loop computational simulation of cardiovascular physiology. Previous PSCOPE methods coupled rigid experiments to a lumped parameter network (LPN) of physiology but are incompatible with volumetrically dynamic experiments where fluid volume varies periodically. We address this limitation by introducing a method capable of coupling multi-branch and volumetrically dynamic in-vitro experiments to an LPN. Methods: Our proposed method utilizes an iterative weighted-averaging algorithm to identify the unique solution waveforms for a given PSCOPE model. We confirm the accuracy of these PSCOPE solutions by integrating mathematical surrogates of in-vitro experiments directly into the LPN to derive reference solutions, which serve as the gold standard to validate the solutions obtained from using our proposed method to couple the same mathematical surrogates to the LPN. Finally, we illustrate a practical application of our PSCOPE method by coupling an in-vitro renal circulation experiment to the LPN. Results: Compared to the reference solution, the normalized root mean square error of the flow and pressure waveforms were 0.001%∼0.55%, demonstrating the accuracy of the coupling method. Conclusion: We successfully coupled the in-vitro experiment to the LPN, demonstrating the real-world performance within the constraints of sensor and actuation limitations in the physical experiment. Significance: This study introduces a PSCOPE method that can be used to investigate medical devices and anatomies that exhibit periodic volume changes, expanding the utility of the hybrid framework.