Statistical Estimation of the Pharmacokinetic Parameters in the One Compartment Open Model.
TEXAS UNIV AT AUSTIN CENTER FOR CYBERNETIC STUDIES
Pagination or Media Count:
A comparative blood-level trail is undertaken to evaluate the in vivo performance of drug formulations in human subjects. A known amount of drug is administered and plasma levels are measured at specified times. The resulting concentration-time curve reflects the absorption, distribution and elimination of the drug. The simplest model that represents this kind of data is the one compartment open model. The one compartment open model is nonlinear in the unknown parameters and is usually fit using least squares. Gradient methods often fail for real world data, and the Nelder-Mead simplex algorithm provides a useful alternative. Further, least absolute deviation criteria can be used as an alternative to least squares. Two numerical examples are presented.