Research Spotlight: Utilizing Differential Evolution into Optimizing Targeted Cancer Treatments

In order to develop an evolvable cancer treatment simulator, the best evolutionary optimisation methods must be found. In this paper our team from University of the West of England and University of Novi Sad, report on using differential evolution to optimise in silico design of a targeted drug delivery system using a PhysiCell simulator. PhysiCell is a multi-cellular agent-based simulator. Differential evolution is used over other well-established evolutionary algorithms because although they use similar algorithmic steps, they are more efficient. They were designed with three main objectives: the ability to find the global optimum, the fast convergence to this optimum and the need for a small amount of control parameters for the procedure. It was found that this was more efficient than a standard genetic algorithm, due to its ability to keep the diversity of the population high. To read more about the results, take a look at the full paper: Utilizing Differential Evolution into Optimizing Targeted Cancer Treatments

The graphical representation output of PhysiCell after 10 days of simulated growth and treatment of a cancer tumour.