MACHINE LEARNING APPROACHES FOR PREDICTING DRUG RELEASE FROM NOVEL DRUG DELIVERY SYSTEMS

Abhishek Ghosh*, Sarmili Sahoo, Piyasa Chakraborty, Chintada Sowjanya, Dr. K. Sudheer Kumar, Seetaramswamy Seepena, Animesh Kumar Tiwari, Indeevar, Saksham Pathak and Kuldeep Singh

This paper aims to present how machine learning methods can be used to predict drug release from novel drug delivery systems, which is one of the most crucial problems in pharmaceutical sciences. The goal therefore is to explain how different machine learning methods can enhance the predictive capabilities for drug release kinetics compared to traditional mechanistic modelling approaches. To achieve this goal, the basic concepts underlying drug release and presentation of data-based approaches are given in detail to provide background knowledge to the interdisciplinary modelling approach. The main outcomes of the paper are that machine learning algorithms make the prediction process easier which also facilitates more tailored formulation design due to the identification of various relations in complex datasets. To conclude, the successful application of machine learning in predictive modelling is presented to have the potential to create disruption in traditional modelling in pharmaceutical science that will lead to more efficient design and optimisation of innovative drug delivery systems.