Modeling and control of anti-canser drug delivery system

Chemotherapy is a widely used method in the treatment of cancer. In chemotherapy treatment, it is essential to determine the optimal drug dose without adversely affecting the cancer patient. Developing technology has enabled the increase in modeling and control studies developed for anti-cancer drug delivery systems. In this thesis, a common patient model that considers cumulative drug toxicity, which can be considered as an important side effect for cancer patients, is used. Using this patient model, different control methods and metaheuristic optimization algorithms are used to achieve optimal chemotherapy drug delivery. Firstly, the drug delivery system is controlled with a fractional-order proportional-integral-derivative (FOPID) controller, which is widely used in the literature. The parameters of the FOPID controller are optimized by five different algorithms: artificial hummingbird algorithm (AHA), gray wolf optimization algorithm (GWO), particle swarm optimization (PSO), salp swarm algorithm (SSA), whale optimization algorithm (WOA). The performance indexes and final number of cancerous cells obtained at the end of the treatment process are compared. In the second part of the study, a new fitness function is generated for the optimization process and applied to the FOPID controlled system. When the final number of cancerous cells and performance indexes are compared, it is seen that the results obtained with the multi-objective optimization of the FOPID controller are further improved. In the last part of the research, a new controller is proposed for the optimal control of the anti-cancer drug delivery system. A hybrid Fuzzy P+Dµ controller is designed by combining a fuzzy logic controller (FLC) and a fractional-order proportional-derivative (FOPD) controller and all parameters of the controller are optimized by PSO algorithm. The performance index and final number of cancerous cells obtained at the end of the treatment process reveal the success of the proposed controller. Robustness tests are also performed for the proposed controller by varying the treatment plan and in-patient parameters. The obtained results are promising for the development of a computer-aided anti-cancer drug delivery system.