Metaheuristic Optimization Algorithms for Enhancing Bergman's Minimal Model

The Bergman Minimum Model (BMM) represents the Glucose-Insulin system. For a better representation of this system, we modify the BMM model by varying its parameters. To validate our Modified Bergman’s Minimal Model (MBMM), we collected Glucose and Insulin data from a normal person. We then reformulated the estimation of MBMM model parameters into a nonlinear optimization problem, which we solved using ten optimization algorithms: the Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Pattern Search Algorithm (PSA), Simulated Annealing Algorithm (SAA), Sequential Quadratic Programming (SQP), Interior Point Method (IPM), Active Set Method (ASM), Firefly Algorithm (FA), Moth Swarm Algorithm (MSA), Stochastic Fractal Search (SFS). The modified BMM model with IPM is more accurate in representing the variation in blood glucose and insulin in a normal person than the BMM model.