A global sensitivity analysis approach applied to a multiscale model of microvascular flow

Area of Science:

• Computational modeling
• Biomedical engineering
• Fluid dynamics

Background:

• Multiscale computational models are essential for simulating complex biological systems like microvascular flow.
• Understanding the sensitivity of these models to various parameters is critical for accurate predictions.
• Vascular network topology is a key structural feature that may influence flow dynamics.

Purpose of the Study:

• To perform a global sensitivity analysis on a multiscale computational model of microvascular flow.
• To identify the influence of input parameters, particularly vascular network topology, on simulation outcomes.
• To quantify the impact of topology on both vasculature-related and interstitium-related variables.

Main Methods:

• Conducted 140 simulations of a multiscale computational model.
• Varied 6 input parameters to assess their effects on 7 output variables.
• Employed global sensitivity analysis to determine parameter importance.

Main Results:

• Vascular network topology emerged as a determinant factor for both vasculature and interstitium quantities.
• Topology significantly influenced average and spatial distribution of vasculature-related outputs (scores 5.5/6 and 6/6).
• Topology also substantially affected average and spatial distribution of interstitium-related outputs (scores 4/6 and 5/6).

Conclusions:

• Vascular network topology plays a critical role in the outcomes of microvascular flow computational models.
• The structural arrangement of the vascular network significantly impacts simulated physiological and interstitial dynamics.
• These findings underscore the importance of accurately representing vascular topology in computational studies.