Tumor Model Fitting using Markov Chain Monte Carlo Method

Accurate mathematical modeling of tumor growth supports understanding treatment effects and experimental planning in preclinical oncology.We apply Markov Chain Monte Carlo (MCMC) methods to estimate parameters of a tumor growth model, defined by ordinary differential equations, using real measurement data from mouse experiments under a predefined dosing schedule.Assuming normally distributed measurement noise, we assess posterior distributions with convergence diagnostics, density plots, and trace plots, while analyzing parameter correlations to evaluate model structure and identifiability.The results highlight that MCMC-based fitting not only provides reliable parameter estimates but also valuable insights into parameter uncertainty, supporting robust modeling strategies in preclinical cancer research.