Rapid prototyping of multi-compartment models for urea kinetics in hemodialysis: a System Dynamics approach.

Models of urea kinetics facilitate a mechanistic understanding of urea transfer and provide a tool for optimizing dialysis efficacy. Dual-compartment models have largely replaced single-compartment models as they are able to accommodate the urea rebound on the cessation of dialysis. Modeling the kinetics of urea and other molecular species is frequently regarded as a rarefied academic exercise with little relevance at the bedside. We demonstrate the utility of System Dynamics in creating multi-compartment models of urea kinetics by developing a dual-compartment model that is efficient, intuitive, and widely accessible to a range of practitioners. Notwithstanding its simplicity, we show that the System Dynamics model compares favorably with the performance of a more complex volume-average model in terms of calibration to clinical data and parameter estimation. Its intuitive nature, ease of development/modification, and excellent performance with real-world data may make System Dynamics an invaluable tool in widening the accessibility of hemodialysis modeling.

Facilitating Understanding, Modeling and Simulation of Infectious Disease Epidemics in the Age of COVID-19.

Interest in the mathematical modeling of infectious diseases has increased due to the COVID-19 pandemic. However, many medical students do not have the required background in coding or mathematics to engage optimally in this approach. System dynamics is a methodology for implementing mathematical models as easy-to-understand stock-flow diagrams. Remarkably, creating stock-flow diagrams is the same process as creating the equivalent differential equations. Yet, its visual nature makes the process simple and intuitive. We demonstrate the simplicity of system dynamics by applying it to epidemic models including a model of COVID-19 mutation. We then discuss the ease with which far more complex models can be produced by implementing a model comprising eight differential equations of a Chikungunya epidemic from the literature. Finally, we discuss the learning environment in which the teaching of the epidemic modeling occurs. We advocate the widespread use of system dynamics to empower those who are engaged in infectious disease epidemiology, regardless of their mathematical background.