Feedback linearization using neural networks applied to advanced pharmacodynamic and pharmacogenomic systems

Abstract

Pharmacological modeling is developing from an empirical discipline into a mechanistic science. Also, new and important fields like pharmacogenomics appeared. As a consequence, pharmacology is dealing with high dimensional, nonlinear, control systems. The intent of this paper is to show that all this systems, being based on a limited array of mechanisms and having some structural peculiarities, are good candidate for the application of feedback linearization techniques, using neural networks. Unlike Jacobian linearization, feedback linearization is not only locally valid. The proposed protocol can be applied even without the aid of a mathematical model. A drug dosage regimen, established in this way, will determine the output of the pharmacological system to track very well the therapeutic objective. To the best of author's knowledge, this is the first time when a very large class of complex pharmacological problems are formulated and solved in terms of neural network control.