Nederlandse Brandwonden Stichting
Using Machine Learning to Speed Up Finite Element Simulations for Wound Care
*Joint project with Ginger Egberts (Hasselt University), Ibrahim Korkmaz, PhD (Amsterdam UMC), Fred Vermolen (Hasselt University), and Paul van Zuijlen (Amsterdam UMC)
Adequate wound-care aims at preventing hypertrophic scars and large contractions that characterize severe burn injuries. To optimize wound-care it is necessary to develop quantitative insight in the underlying biological mechanisms occurring during post-burn skin evolution. In earlier projects, a mathematical foundation for the modelling of post-burn skin evolution has been laid. This foundation is based on a set of nonlinearly coupled partial differential equations (PDEs), solved using finite elements. Since many of the input parameters are patient-specific or badly documented, the simulations contain uncertainty. Hence simulation outcomes are only useful in a Bayesian sense, where one predicts the probability distribution over various scenarios. Hence multiple finite element runs are required, which makes the finite element method unattractive in clinical settings. Therefore, we will develop an efficient computational framework that reproduces the expensive finite element simulations. This tool helps clinicians predict the probability of success for various scenarios and treatments.