Model based design of cancer chemotherapy drug scheduling: a particle swarm optimization approach

Abstract

This paper presents an investigation into the development of a cancer model for chemotherapy drug scheduling using Particle Swarm Optimization (PSO) algorithm. PSO is a population-based search method whose mechanics are inspired by the ability of flocks of birds, schools of fish, and herds of animals to adapt to their environment and find rich sources of food by implementing “information sharing” approaches. The main aim of chemotherapy treatment is to reduce the tumor size to a desired minimum level so that cannot be detected in vivo clinically. 'Mouse' and 'Human' models developed by de Pillis and co-workers in 2006 are used to design chemotherapy drug doses and observe its effects on different cell populations. Besides chemotherapy, these models are also used to study the effects of immunotherapy, anti-angiogenic therapy or combinations of these. Chemotherapy drug scheduling is designed as optimal control problem based on these models and PSO is used to find drug doses for specific intervals and periods relevant to clinical practice. Results show that the employed method can generate a wide range of solutions that trade-off between cell killing and toxic side effects and satisfy associated goals of chemotherapy treatment.