FDA-collaboration

This project is motivated by the goal of elevating computational modeling and simulation from a scientific research tool to a high-quality regulatory tool, in response to priorities set by the U.S. Food and Drug Administration. It adopts and expands the framework of computational reproducibility, proposing that regulatory grade computational evidence hinges on unimpeachable provenance. This means capturing, documenting and managing the full iterative workflow, decision making, computational artifacts (data and code), and resulting evidence with maximum transparency. Through a case study, the project develops best practices for constructing computational evidence in support of regulatory submissions. The case consits of an electronic drug delivery system that derives from e-cigarette devices, and the study develops a computational model of the fluid flow in the device. In the process, credibility building activities are documented, including code and solution verification, following: ASME Standard, V&V 40–2018, Assessing Credibility of Computational Modeling through Verification and Validation: Application to Medical Devices. The model was developed using the open source software OpenFOAM, and was based on preliminary FDA studies that established the risk, context of use, and question of interest to motivate device geometry and operaating conditions. Using sensitivity analysis, better understanding was gained of the relationship between inputs and outputs. A risk-informed credibility analysis was conducted based on the US FDA Guidance, ASME V&V 20 Standard, the ASME V&V 40 Standard, and the Credibility Goals established by the preliminary study. A PhD student was funded for one year to conduct the project, and the work was presented via posters at two research events.