IDENTIFIABILITY AND INDISTINGUISHABILITY OF NONLINEAR PHARMACOKINETICMODELS

Three nonlinear model structures of interest in pharmacokinetics are a nalyzed to determine whether the unknown, independent, model parameter s can be deduced if perfect input-output data were available. This is the problem of identifiability. The method used is based on the local state isomorphism theorem. In certain circumstances, the modeler may b e undecided between several model structures and it is then of interes t to determine whether different model structures can be distinguished from perfect input-output data. This is the problem of model indistin guishability. The technique used, again based on the local state isomo rphism theorem, parallels the similarity transformation approach for l inear systems described previously in this journal. The analysis is pe rformed on three two-compartment examples having one linear and one no nlinear (Michaelis-Menten) elimination pathway. In each model there is , on physiological and other grounds, some uncertainty over the precis e location (central compartment or peripheral compartment) of one of t he elimination pathways.