RESEARCH ACTIVITIES
Current research activities include: application of stochastic control theory to the design of optimal dosage regimens; improved methods for population pharmacokinetic modeling; investigation of process and measurement noise in the clinical environment; evaluation of alternative parameterizations of pharmacokinetic models; optimal monitoring strategies for pharmacokinetic studies, clinical trials, and patient care; and development of clinically reliable interfaces and apparatus for delivering complex drug infusion regimens.
RESEARCH FACILITIES
Software is available for use in pharmacokinetics research applications. Current computer programs include:
- BOXES: A program package for simulation and parameter estimation, in linear and non-linear pharmacokinetic systems. Outputs model files in the form of Fortran source code for the differential equations describing the system.
- The MM-USCPACK clinical package: A PC program package for clinical use in planning, monitoring, and adjusting dosage regimens of digoxin, aminoglycoside antibiotics, vancomycin, and other drugs.
- The IT2Band NPEM2 (NPAG): Programs for Population Modeling of Drug Behavior. They use aniterative 2-stage Bayesian method and a nonparametric adaptive grid algorithm for computing the discrete joint density of the parameters in PK/PD models.
- MMLQ2: A Multiple Model Linear Quadratic stochastic control program which designs dosage regimens to minimize a therapeutic cost function, thus optimizing therapeutic precision about a specific selected therapeutic goal. It takes into account arbitrary process and measurement uncertainties inherent in the clinical environment of drug therapy. It uses the NPAG population model for the Bayesian prior.
- Maximum Entropy Program: Develops a discrete joint density from data of parameter means and variances, for use with the MMLQ dosage designer.
- MC-SIM: A package Monte Carlo Simulator for both parametric and nonparametric PK/PD models.
COLLABORATIVE AND SERVICE ACTIVITIES
The software and facilities can be made available, with approval, to researchers for collaborative projects or as a technical service.

