Using the stochastic collocation method for the uncertainty quantification of drug concentration due to depot shape...

Area of Science:

• Computational modeling and simulation
• Biomedical engineering
• Pharmacokinetics

Background:

• Numerical simulations rely on modeling assumptions that influence results.
• Quantifying the probabilistic relationship between model variability and outcome variability is essential for reliable predictions.
• Traditional Monte Carlo methods can be computationally intensive for assessing parameter variability.

Purpose of the Study:

• To develop a framework for studying the impact of drug depot shape variability on drug diffusion simulations.
• To assess the efficacy of the stochastic collocation method as an alternative to Monte Carlo approaches in this context.
• To integrate component shape parameterization with stochastic collocation for realistic simulations.

Main Methods:

• Utilized realistic geometries from MR images of a porcine model.
• Employed level-set techniques for creating univariate shape parameterizations.
• Applied the stochastic collocation method to model variability and outcome.

Main Results:

• Demonstrated a computationally feasible approach to assess the effect of shape variability.
• Showcased the straightforward quantification of variability once the stochastic process is characterized.
• Provided a method for validating and verifying simulation outcomes influenced by shape parameters.

Conclusions:

• The proposed framework effectively studies drug depot shape variability's impact on diffusion simulations.
• Stochastic collocation offers an efficient alternative for assessing model and parameter variability.
• This approach is a significant step towards robust validation and verification in computational modeling.