Phase specific optimal treatment for cancer using GA and swarm intelligence

This paper presents an investigation into the development of a multi-objective optimal chemotherapy control model to reduce the number of cancer cells after a number of fixed treatment cycles with minimum side effects. A phase specific drug scheduling method using a close-loop control method with multi-objective techniques is proposed in this paper. Genetic Algorithm (GA) and particle swarm optimisation algorithm (PSO) are used to optimise the control solution for trading-off between the cell killing and toxic side effects. A close-loop control method, namely Integral-Proportional-Derivative (I-PD) is designed to control the drug to be infused into the patient's body and multi-objective GA (MOGA) and multi-objective PSO (MOPSO) are used to find suitable parameters of the controller. The proposed algorithm is implemented, tested and verified through a set of experiments. Performances of the proposed methods demonstrated that both the MOGA and MOPSO approach can offer very efficient drug scheduling that trade-off between cell killing and toxic side effects and satisfy associated design goals. It is also noted that the MOGA based method offers better performance as compared to MOPSO and can reduce the number of proliferating and quiescent cells up to 72.2% and 60.4% respectively. Future research needs to evaluate the proposed scheduling with clinical data and experiments.