A cancer-inspired metaheuristic for optimal control of breast cancer treatment

This paper introduces a novel bio-inspired metaheuristic, referred to as the Cancer Improvement Optimizer (CIO), which is motivated by the fundamental biological processes underlying cancer progression, including tumor proliferation, immune surveillance, therapeutic intervention, dormancy, and recurrence. By emulating the dynamic interactions among these mechanisms, CIO establishes an effective balance between global exploration and local exploitation, thereby enhancing convergence accuracy while reducing the risk of premature convergence to local optima. The performance of the proposed CIO algorithm is first assessed on a comprehensive collection of continuous benchmark functions encompassing unimodal, multimodal, separable, and non-separable optimization problems. Extensive numerical comparisons with several state-of-the-art metaheuristic algorithms demonstrate that CIO consistently obtains competitive or superior solution quality, faster convergence, and greater robustness across a wide range of Benchmark functions while maintaining satisfactory computational efficiency. To further demonstrate its practical applicability, CIO is employed to solve an optimal control problem (OCP) for breast cancer treatment. The proposed mathematical framework combines chemotherapy, immunotherapy, anti-cancer drug administration, and ketogenic diet intervention within a nonlinear system of ordinary differential equations describing tumor-immune interactions. The resulting OCP is discretized using a collocation scheme based on Boubaker scaling functions, yielding a finite-dimensional nonlinear programming problem that is efficiently solved using CIO. Numerical results show that the proposed hybrid framework effectively suppresses tumor growth, preserves healthy-cell populations, strengthens the immune response, and generates adaptive treatment strategies that achieve an appropriate balance between therapeutic efficacy and treatment-related toxicity. These findings demonstrate that CIO is not only an effective general-purpose optimization algorithm but also a reliable computational framework for solving complex biomedical OCPs.