MACHINE LEARNING IN PHARMACEUTICAL FORMULATION DEVELOPMENT

Abstract

The use of ML algorithms in pharmaceutical research has enabled remarkable progress, especially by circumventing the problems inherent in the trial and error method of drug development. This review is intended to highlight the growing use of machine learning algorithms such as artificial neural network models, random forest models, and support vector machines in the optimization of pharmaceutical formulations. Using machine learning algorithm, a complex set of information may be analyzed to predict the rate of drug release, stability, and other physical and chemical properties, hence cutting down the time required and saving costs in development process. Additionally, ML algorithms allow formulation of innovative systems as well as optimization of processes. In spite of the issues of insufficient data, lack of model transparency, and regulations, it is expected that future innovations in artificial intelligence and computation will solve these problems.