Model Predictive Control of Melanoma Treatment Enhanced by Particle Swarm Optimization

This study analyzes the growth dynamics of melanoma tumor cells and develops a model predictive controller (MPC) using four well-known optimizers to suppress tumor growth, proposing an MPC framework that integrates multiple metaheuristic algorithms for regulating tumor size. All modelling, control design, and simulations are performed in MATLAB, and results indicate that a PSO-based MPC offers satisfactory response and rapid convergence, achieving effective tracking and disturbance rejection. The study assumes precise drug dosing is feasible and demonstrates substantial tumor-size reduction through the integration of MPC with metaheuristic optimization. Simulation findings reveal that the PSO-based MPC achieves notable improvement in tumor reduction and overall control performance, outperforming other metaheuristic approaches, as evidenced by comparative error metrics: ITAE ≈ ۱.۹۳۷۷ × ۱۰^۳, IAE ≈ ۲۴۴.۴۵, MSE ≈ ۴.۶۸۶۳ × ۱۰^۳.

کلیدواژه ها:

Model predictive control‎ ، ‎Melanoma cancer‎ ، ‎Tumor growth control ، Biomedical control systems ، ‎Metaheuristic optimization