An efficient iteration procedure for the cluster newton method in inverse parameter identification of pharmacokinetics

In this paper, we proposed an efficient iteration procedure, being used in the initial cluster Newton method (CNM) to seek multiple solutions for inverse parameter identification in pharmacokinetics simultaneously. Instead of the constant level of perturbation in the initial CNM, the reasonable level of perturbation is controlled corresponding to different iterations. The simulation results of the proposed scheme have pointed out that we can reduce iterations and computational time; the point cluster goes more stably towards the solution manifold. © 2020 IEEE.