ITSITS

(IJCSAM) International Journal of Computing Science and Applied Mathematics(IJCSAM) International Journal of Computing Science and Applied Mathematics

An alternative method of integration-based parameter estimation applied in pharmacokinetics problems is proposed here. The method, introduced by Holder and Rodrigo [1], is used to estimate the rate of drug elimination and distribution when it enters the body via intravenous bolus. The estimation results are then compared with the classical method, the least squares method for the one-compartment model, and the residual method for the two-compartment model. Graphical simulations of drug concentration versus time are also performed in this article to view not only the dynamics of drug delivery in the body, but also the comparisons between the approximate solutions and the arbitrarily generated data points. Comparisons are also presented when the data points take into account noise in the form of random values. Based on the estimation and simulation results, the integration-based method gives good results and even better than the classical method although when noise is applied to the data points.

The integration-based method presents a viable alternative for estimating parameters in pharmacokinetic models, specifically for one and two-compartment systems.This method demonstrates comparable precision to the classical least squares method in the one-compartment model.Furthermore, the integration-based method provides superior parameter estimates compared to the classical residual method in the two-compartment model, particularly when dealing with noisy data.

Future research should explore the application of the integration-based method to more complex pharmacokinetic models, such as those incorporating multiple compartments or non-linear elimination kinetics, to assess its robustness and scalability. Investigating the sensitivity of the method to different types and levels of noise in the data could further refine its practical implementation and improve its reliability in real-world scenarios. Additionally, a comparative study with other advanced parameter estimation techniques, like machine learning algorithms, would provide valuable insights into the strengths and limitations of the integration-based approach and potentially lead to hybrid methodologies that leverage the benefits of both.

Read online
File size295.4 KB
Pages6
DMCAReport

Related /

ads-block-test