A Pathwise Fractional one Compartment Intra-Veinous Bolus Model

Abstract

Extending deterministic compartments pharmacokinetic models as diffusions seems not realistic on biological side because paths of these stochastic processes are not smooth enough. In order to extend one compartment intra-veinous bolus models, this paper suggests to modelize the concentration process $C$ by a class of stochastic differential equations driven by a fractional Brownian motion of Hurst parameter belonging to $]1/2,1[$. The first part of the paper provides probabilistic and statistical results on the concentration process $C$ : the distribution of $C$, a control of the uniform distance between $C$ and the solution of the associated ordinary differential equation, an ergodic theorem for the concentration process and its application to the estimation of the elimination constant, and consistent estimators of the driving signal's Hurst parameter and of the volatility constant. The second part of the paper provides applications of these theoretical results on simulated concentration datas : a qualitative procedure for choosing parameters on small sets of observations, and simulations of the estimators of the elimination constant and of the driving signal's Hurst parameter. The relationship between the estimations quality and the size/length of the sample is discussed.