A hybrid intelligent system for formulation of BCS Class II drugs in hard gelatin capsules

Abstract

In this paper, we describe a hybrid intelligent system for formulation of BCS Class II drugs in hard gelatin capsules. Several significant challenges are involved in drug-formulation: the active ingredients and the fillers in the capsule must be chemically compatible according to bio-pharmaceutical principles; the formulation must be manufacturable; and it must meet the prescribed drug release requirement. Traditional trial and error approach to drug-formulation design is too costly and time consuming to meet the increasing demand for new drugs. To answer these challenges, we have developed a prototype hybrid intelligent system for automatic drug formulation. This system consists of a rule-based Expert System (ES) that conducts formulation design according to Biopharmaceutical Classification of drugs and a neural network (NN) that predicts the quality of the formulation recommended by ES. Through interaction between the two modules, the hybrid system forms a (re) formulation-prediction cycle, and the quality of the formulation is improved with each iteration. The hybrid system is tested with sample drugs and is shown to be able to produce formulations with desirable performance measures.