Evaluating the accuracy of cerebrovascular computational fluid dynamics modeling through time-resolved experimental...

Area of Science:

• Biomedical Engineering
• Fluid Dynamics
• Neuroscience

Background:

• Accurate modeling of cerebral hemodynamics is vital for understanding stroke.
• Computational fluid dynamics (CFD) is a key tool for cerebrovascular research.
• A comprehensive accuracy assessment of CFD models with transient pressures and flows is lacking.

Purpose of the Study:

• To systematically evaluate the accuracy of different outlet boundary conditions (BCs) in cerebrovascular CFD models.
• To compare CFD modeling results against in-vitro experimental data.
• To identify limitations and provide recommendations for improving CFD accuracy in cerebrovascular research.

Main Methods:

• An in-vitro experimental setup using an anatomical cerebrovascular phantom was created.
• High-resolution flow and pressure data were acquired from the phantom.
• CFD models with five sets of stationary and transient BCs, including a novel phase modulation approach, were developed and compared to experimental data.

Main Results:

• The experiment yielded physiological hemodynamics consistent with clinical data.
• In-silico models exhibited significant time-dependent deviations: 19-66% for flows and 6-26% for pressures.
• Stationary outlet pressure BCs resulted in the highest deviations, while Windkessel and phase modulation BCs showed promise for specific parameters.

Conclusions:

• Current cerebrovascular CFD models demonstrate limited accuracy, particularly with stationary outlet pressure BCs.
• The Windkessel and phase modulation BCs offer improvements for flow pulsatility and pressure, respectively.
• Further research is needed to enhance the accuracy and reliability of CFD models for cerebrovascular applications.