In silico chemotherapy optimization with genetic algorithm

The combination of medicine with engineering has great potential. The currently used chemotherapy treatments usually use maximal tolerable doses, which can lead to harmful side effects. By using a mathematical approach, we are able to personalize chemotherapy treatments, using unique patient parameters. We propose an algorithm that is capable of generating a chemotherapy treatment plan to cure cancer patients. The objective of the algorithm is to create a treatment that shrinks the tumor while minimizing the injected doses to decrease treatment costs and prevent drug toxicity. In this paper, we used a genetic algorithm to find the optimal treatment. First, we optimized the therapy on a single patient, later we carried out therapy optimization on a population with predefined ranges for the patient model parameters. The parameters are acquired from in vivo mice experiments through parametric identification. According to the results, the generated treatment produced higher survival rates with slightly higher doses compared to the standard clinically used treatment.